1011

Int J Environ Res 9(3)1011-1022 Summer 2015ISSN 1735-6865

Received 19 Sep 2014 Revised 9 Dec 2014 Accepted 13 Dec 2014

Corresponding author E-mail mzmoceangmailcom

Survival of High Latitude Fringing Corals in Extreme TemperaturesRed Sea Meteorology

Moustafa MZ1 Moustafa ZQ2 Moustafa MS1 Moustafa SE3 and Moustafa ZD4

1H2O amp Geo Inc 70 Whispering Oaks Trail West Palm Beach FL 33411 USA2Wilson College Princeton University 110 West College PO Box 430 Princeton NJ 08542-0430

3Department of Geography Rutgers The State University of New Jersey Piscataway NJ 08854-80454 Pratt School of Engineering Duke University Durham NC 27708

ABSTRACT Zakirsquos Reef is located in the Gulf of Suez a narrow portion of the Red Sea with exceptionally dryand hot climate and lacks almost all scientific data This research intends to describe the arearsquos unique climatologywhich may reveal correlations between the reefrsquos existence at high latitude and extreme climate conditions Airtemperature at Zakirsquos Reef fluctuated between 03 and 586oC with a daily range of 25oC and annual mean of221oC plusmn0075 Spectral analysis revealed half day daily and yearly return periods all of which have shown thatdaily and half daily cycles are dominating the local climate with amplitudes of approximately 5oC Frequencyhistograms revealed a bimodal signal one peak at 14-15oC and a second peak at 32-33oC both represent nighttimeand daytime temperatures half-day cycle Meteorological data collected at Zakirsquos Reef were also compared toHurghadarsquos and Ismailiarsquos 400 and 200 km south and north of the study site respectively to reveal any anomalies Although air temperature daily means at Zakirsquos Reef were similar to Hurghadarsquos maximum daily means exceededHurghadarsquos by 7oC while minimums were almost equal to Ismailiarsquos Unexpected temperature trends and shortdistance between mountain range and Zakirsquos Reef prompted us to hypothesize that a Foehn wind may beresponsible for the extremely high air temperatures Air parcel trajectory model results further verified that localwind patterns matched signatures of a Foehn wind The observed warmer than normal air temperatures may beresponsible for securing the survival of these northernmost subtropical coral reefs

Key words Red Sea Gulf of Suez Fringing Reef Air Temperature Foehn Wind

INTRODUCTIONThe Red Sea is truly one of the worldrsquos most exotic

and fascinating natural environments Located betweenAsia and Africa its most northern point forms the SinaiPeninsula and stretches about 1000 miles south to jointhe Indian Ocean The Red Sea is home to more thanone thousand species of fish and over four hundredspecies of coral more than any other proportionallysized body of water (Hanaur 1988) and remains at analmost constant temperature averaging 22oC yearround

Zakirsquos Reef (the study site) is an isolated fringingcoral reef that extends approximately one-kilometerparallel to shore and lies 50-100 m offshore (Figure 1)and is located in the Red Searsquos Gulf of Suez near themost northern latitude for subtropical coral reefs (29o

32rsquoN and 32o 24rsquoE) The very shallow (50-80m deep)Gulf of Suez lies adjacent to an extremely arid desertwhere rainfall is minimal evaporation rates are highand freshwater inputs are non-existent The persistentNorthwesterly trade winds and extreme temperaturesresult in high salinity (43-45 psu) as well as large dailyand seasonal temperature fluctuations Hightemperatures and salinity reduce oxygen solubilitywhich can stress many species of reef-dwellingorganisms The organisms that thrive here must beable to tolerate these environmental extremes Of theestimated 335 species of corals found in the Red Seaonly 35 species have been identified in the Gulf ofSuez (Mustafa 2000) and of these only six specieshave been identified to dominate Zakirsquos Reef(Moustafa et al 2008a) accounting for more than 95of total coral coverage

1012

Moustafa MZ et al

The Gulf of Suez is a narrow portion of the RedSea surrounded on all sides by mountain rangesAlmost all forms of scientific data are lacking for thisregion (Moustafa et al 2008a b) Therefore theoverarching goal of this multi-year study was toestablish a knowledge base for fringing coral reefs inthe Gulf of Suez and provide basic information forcurrent and future reef management as reef-relatedactivities generate a significant portion of the countryrsquostourism revenue accounting for approximately 11 ofEgyptrsquos GNP and providing 22 million jobs (Hanaur1988 UNESCO 2005)

In situ observations from the study site and twoadditional stations located north and south of the studysite spanning over a two year period (2007ndash2009) werecollected and analyzed to describe prevailingconditions and reveal any existing anomalies in thisregionrsquos climatology Long-term data collectionprovides evidence of seasonal and interannualvariability and may reveal correlations between atropical reefrsquos existence at such high latitude andclimate conditions of this region Specific objectivesof this research were

1 Establish observed meteorological data ranges andtrends at the northern limits of sub-tropical coral reefecosystems the nearest available historical weatherstation is more than 200 km away

2 Determine the cyclic nature of all univariate timeseries to reveal natural forcesphenomena that exist inthis region and

3 Compare our results to other similar studies andlocalities which may explain and identify the most likelyreasons why this fringing coral reef survives at thenorthernmost latitude of subtropical reefs ieinvestigate if the local climate is responsible as a wholeor in part for this reefrsquos survivalship

MATERIALS amp METHODSA roof top weather station was placed near Zakirsquos

Reef at a height of ten meters above sea level Highfrequency time series of air temperature dew pointand humidity were collected with highly sensitivesensors for two years between August 20 2007 andSeptember 16 2009 (Onset Computer Corporation2014)

We were also able to obtain daily-meanobservations from two nearby stations Ismailia andHurghada located 200 and 400km away north and southof the study site respectively (Wunderground 2010)Daily data included air temperature dew point andrelative humidity at these locations (Figure 1)

A combination of descriptive statistics togetherwith graphical techniques was employed to identifyseasonality or trends and describe relationshipsbetween air temperatures and local climate near ZakirsquosReef All analyses were performed with daily monthlyand yearly means calculated from 30 minuteobservations (48 samples per day) Statistics werecalculated with JMP (Version 7 SAS Institute Inc CaryNC) or SigmaPlot (Version 110 SPSS Inc ChicagoIL) The level of significance (α) was set at 005 for all

Fig 1 Location of Zakirsquos Reef in the Red Searsquos Gulf of Suez approximately 80 km south of the highlytrafficked Suez Canal Panel A general location of the study area Zakirsquos reef is located approximately 60 km

from Suez City 200 km south of Ismailia and 400 km north of Hurghada Panel B Location of Zakirsquos reefrelative to Gabel Ataka mountain (distance is lt 500 meters)

A B

1013

Int J Environ Res 9(3)1011-1022 Summer 2015

analyses Box and whisker plots were also developedand summarized

Spectral analysis was used as the primarytechnique for assessing the cyclic nature of allunivariate time series in the frequency domain andwas applied to all time series after filtering the originaldata with a 30 hour filter to exclude high frequencysignals Spectral analysis was applied to the rawdatasets with no filtering and for the original half hourlyand hourly observations Spectral analysis was alsoapplied to a new generated dataset that was createdby subtracting the overall mean from the original half-hourly observations

HYSPLIT model was used to investigate localcirculation patterns Wind-patterns were modeled usingthe National Ocean and Atmospheric Associationrsquos(NOAArsquos) HYSPLIT (HYbrid Single-Particle LagrangianIntegrated Trajectory) model (Draxler and Rolph 2013)The HYSPLIT model provides a variety of differentforms of quantitative information such asmeteorological topographical and environmentalfactors involved in the regionrsquos weather patterns Thismodel was used to investigate trends and patternsregarding the movement of wind and weather aroundthe Gulf of Suez region Model run input data werebased on archived meteorological data trajectoriesextracted from the National Center for EnvironmentalPrediction (NCEP) Global Data Assimilation System(GDAS) datasets available from 2006-present (Okamotoand Derber 2006) The HYSPLIT model runs werebased on single starting point trajectories Pointlocations were selected based on regions of knowndominant wind patterns (ie trade winds) and exactboundaries between mean sea level and Gebel AtakaMountain (located approximately lt 500 m west of ZakirsquosReef) derived from a Digital Elevation Model (DEM)Identifying precise topographic variations in the regionwas crucial in order to distinguish model-predicted airmass spatio-temporal patterns which are formed as aresult of orthographic uplift

RESULTS amp DISCUSSIONObserved air temperatures in this arid environment

reached as high as 59oC in a single day yet the annualmeans were 221 and 256oC for first and secondsampling year respectively (Table 1) Observedminimums were as low as 03 and 33oC while maximumvalues reached as high as 507 and 586oC for 2008 and2009 respectively (Table 1) These wide temperaturefluctuations translate to an average daily range inexcess of 25degC

Overall air temperature means for the two yearperiod were 223oC (plusmn022 SE) for Ismailia 242oC (plusmn024SE) at the study site and 252oC (plusmn022 SE) forTa

ble 1

Dai

ly d

escr

iptiv

e sta

tistic

s for

air

tem

pera

ture

hum

idity

and

dew

poi

nt te

mpe

ratu

re d

ata

colle

cted

at t

he st

udy

site

from

Aug

ust 2

1 2

007

thro

ugh

June

20 2

008

and

from

Jun

e 20

200

8 th

roug

h Se

ptem

ber 1

6 2

009

All

daily

stat

istic

s are

bas

ed o

n 48

dai

ly o

bser

vatio

ns

1014

Survival of Fringing Corals in Extreme Temperatures

Hurghada (Table 2) Air temperature daily means atthe study site were similar to daily means observed atHurghada (south of the study by approximately 400km)yet lower than Ismailia

Relative humidity daily means at the study sitewere significantly higher compared to values from theother two sites Their means were 724 (plusmn013 SE) 168(plusmn018 SE) and 55 (plusmn022 SE) for Ain Sokhna Ismailia

and Hurghada respectively (Fig 2) Dew point dailymeans were very close to observed daily meantemperatures at the study site and higher than thoserecorded at the two other sites (Fig 2)

Dew point daily means were 172 (plusmn024 SE) 135(plusmn020 SE) and 107oC (plusmn018 SE) for Ain SokhnaIsmailia and Hurghada respectively (Fig 2) Airtemperature monthly means had a well-defined annual

Table 2 Statistical summary of daily air temperatures at Ain Sokhna Hurghada and Ismailia from 08212007 through 09162009 Study site is located 200km south of Ismailia and 400km north of Hurghada Daily

observations at study site are based on half-hourly sampling frequency while daily means for the remainingtwo nearby stations were reported and obtained from httpwwwwundergroundcomglobalEGhtml

Ain Sokhna Hurghada Ismailia1 Mean Max Min Mean Max Min Mean Max Min

Mean 2422 3690 1606 2520 3002 2039 2229 2820 1639 Standard Error 024 028 023 022 022 023 022 024 022

Standard Deviation 672 773 630 607 599 636 609 655 590 Median 2506 3744 1676 2556 3056 2056 2333 2889 1667 Mode 1755 4152 2401 3111 3667 2778 3000 3500 2389

Minimum 957 1752 029 1167 1556 167 889 1167 056 Maximum 3644 5855 2712 3556 4167 3056 3333 4278 2667

Count 758 758 758 758 758 758 747 747 747

Fig 2 Daily mean temperature relative humidity and dew points at study site (Ain Sokhna) Hurghada andIsmailia from August 21 2007 through September 16 2009 Solid line air temperature dotted line relative

humidity and dashed line dew point temperature Mean values for the entire period of record

1 = Missing a total of 11 points 8 beginning of May 2009 2 beginning of August 2009 and 1 July 12 2009

1013

Int J Environ Res 9(3)1011-1022 Summer 2015

analyses Box and whisker plots were also developedand summarized

Spectral analysis was used as the primarytechnique for assessing the cyclic nature of allunivariate time series in the frequency domain andwas applied to all time series after filtering the originaldata with a 30 hour filter to exclude high frequencysignals Spectral analysis was applied to the rawdatasets with no filtering and for the original half hourlyand hourly observations Spectral analysis was alsoapplied to a new generated dataset that was createdby subtracting the overall mean from the original half-hourly observations

HYSPLIT model was used to investigate localcirculation patterns Wind-patterns were modeled usingthe National Ocean and Atmospheric Associationrsquos(NOAArsquos) HYSPLIT (HYbrid Single-Particle LagrangianIntegrated Trajectory) model (Draxler and Rolph 2013)The HYSPLIT model provides a variety of differentforms of quantitative information such asmeteorological topographical and environmentalfactors involved in the regionrsquos weather patterns Thismodel was used to investigate trends and patternsregarding the movement of wind and weather aroundthe Gulf of Suez region Model run input data werebased on archived meteorological data trajectoriesextracted from the National Center for EnvironmentalPrediction (NCEP) Global Data Assimilation System(GDAS) datasets available from 2006-present (Okamotoand Derber 2006) The HYSPLIT model runs werebased on single starting point trajectories Pointlocations were selected based on regions of knowndominant wind patterns (ie trade winds) and exactboundaries between mean sea level and Gebel AtakaMountain (located approximately lt 500 m west of ZakirsquosReef) derived from a Digital Elevation Model (DEM)Identifying precise topographic variations in the regionwas crucial in order to distinguish model-predicted airmass spatio-temporal patterns which are formed as aresult of orthographic uplift

RESULTS amp DISCUSSIONObserved air temperatures in this arid environment

reached as high as 59oC in a single day yet the annualmeans were 221 and 256oC for first and secondsampling year respectively (Table 1) Observedminimums were as low as 03 and 33oC while maximumvalues reached as high as 507 and 586oC for 2008 and2009 respectively (Table 1) These wide temperaturefluctuations translate to an average daily range inexcess of 25degC

Overall air temperature means for the two yearperiod were 223oC (plusmn022 SE) for Ismailia 242oC (plusmn024SE) at the study site and 252oC (plusmn022 SE) forTa

ble 1

Dai

ly d

escr

iptiv

e sta

tistic

s for

air

tem

pera

ture

hum

idity

and

dew

poi

nt te

mpe

ratu

re d

ata

colle

cted

at t

he st

udy

site

from

Aug

ust 2

1 2

007

thro

ugh

June

20 2

008

and

from

Jun

e 20

200

8 th

roug

h Se

ptem

ber 1

6 2

009

All

daily

stat

istic

s are

bas

ed o

n 48

dai

ly o

bser

vatio

ns

1014

Survival of Fringing Corals in Extreme Temperatures

Hurghada (Table 2) Air temperature daily means atthe study site were similar to daily means observed atHurghada (south of the study by approximately 400km)yet lower than Ismailia

Relative humidity daily means at the study sitewere significantly higher compared to values from theother two sites Their means were 724 (plusmn013 SE) 168(plusmn018 SE) and 55 (plusmn022 SE) for Ain Sokhna Ismailia

and Hurghada respectively (Fig 2) Dew point dailymeans were very close to observed daily meantemperatures at the study site and higher than thoserecorded at the two other sites (Fig 2)

Dew point daily means were 172 (plusmn024 SE) 135(plusmn020 SE) and 107oC (plusmn018 SE) for Ain SokhnaIsmailia and Hurghada respectively (Fig 2) Airtemperature monthly means had a well-defined annual

Table 2 Statistical summary of daily air temperatures at Ain Sokhna Hurghada and Ismailia from 08212007 through 09162009 Study site is located 200km south of Ismailia and 400km north of Hurghada Daily

observations at study site are based on half-hourly sampling frequency while daily means for the remainingtwo nearby stations were reported and obtained from httpwwwwundergroundcomglobalEGhtml

Ain Sokhna Hurghada Ismailia1 Mean Max Min Mean Max Min Mean Max Min

Mean 2422 3690 1606 2520 3002 2039 2229 2820 1639 Standard Error 024 028 023 022 022 023 022 024 022

Standard Deviation 672 773 630 607 599 636 609 655 590 Median 2506 3744 1676 2556 3056 2056 2333 2889 1667 Mode 1755 4152 2401 3111 3667 2778 3000 3500 2389

Minimum 957 1752 029 1167 1556 167 889 1167 056 Maximum 3644 5855 2712 3556 4167 3056 3333 4278 2667

Count 758 758 758 758 758 758 747 747 747

Fig 2 Daily mean temperature relative humidity and dew points at study site (Ain Sokhna) Hurghada andIsmailia from August 21 2007 through September 16 2009 Solid line air temperature dotted line relative

humidity and dashed line dew point temperature Mean values for the entire period of record

1 = Missing a total of 11 points 8 beginning of May 2009 2 beginning of August 2009 and 1 July 12 2009

1014

Survival of Fringing Corals in Extreme Temperatures

Hurghada (Table 2) Air temperature daily means atthe study site were similar to daily means observed atHurghada (south of the study by approximately 400km)yet lower than Ismailia

Relative humidity daily means at the study sitewere significantly higher compared to values from theother two sites Their means were 724 (plusmn013 SE) 168(plusmn018 SE) and 55 (plusmn022 SE) for Ain Sokhna Ismailia

and Hurghada respectively (Fig 2) Dew point dailymeans were very close to observed daily meantemperatures at the study site and higher than thoserecorded at the two other sites (Fig 2)

Dew point daily means were 172 (plusmn024 SE) 135(plusmn020 SE) and 107oC (plusmn018 SE) for Ain SokhnaIsmailia and Hurghada respectively (Fig 2) Airtemperature monthly means had a well-defined annual

Table 2 Statistical summary of daily air temperatures at Ain Sokhna Hurghada and Ismailia from 08212007 through 09162009 Study site is located 200km south of Ismailia and 400km north of Hurghada Daily

observations at study site are based on half-hourly sampling frequency while daily means for the remainingtwo nearby stations were reported and obtained from httpwwwwundergroundcomglobalEGhtml

Ain Sokhna Hurghada Ismailia1 Mean Max Min Mean Max Min Mean Max Min

Mean 2422 3690 1606 2520 3002 2039 2229 2820 1639 Standard Error 024 028 023 022 022 023 022 024 022

Standard Deviation 672 773 630 607 599 636 609 655 590 Median 2506 3744 1676 2556 3056 2056 2333 2889 1667 Mode 1755 4152 2401 3111 3667 2778 3000 3500 2389

Minimum 957 1752 029 1167 1556 167 889 1167 056 Maximum 3644 5855 2712 3556 4167 3056 3333 4278 2667

Count 758 758 758 758 758 758 747 747 747

Fig 2 Daily mean temperature relative humidity and dew points at study site (Ain Sokhna) Hurghada andIsmailia from August 21 2007 through September 16 2009 Solid line air temperature dotted line relative

humidity and dashed line dew point temperature Mean values for the entire period of record

1 = Missing a total of 11 points 8 beginning of May 2009 2 beginning of August 2009 and 1 July 12 2009

1015

Int J Environ Res 9(3)1011-1022 Summer 2015

Fig 3 Monthly Box and whisker plots for temperature for Ain Sokhna from August 21 2007 throughSeptember 16 2009 top and bottom of box = 75th 25th percentiles respectively black solid midline in box =

median dashed midline in box = mean values ends of whiskers denote 5thand 95th respectively

Table 3 Yearly descriptive statistics for meteorological data collected at the study site for year 2007-20082008-2009 (only matching dates for previous year ie equal number of observations) and 2007-2009 All

statistical values are calculated based on 48 samples per day

2007-2008 Mean Temp

(oC)

Minimum Temp

(oC)

Maximum Temp (oC)

Mean Humidity

()

Minimum Humidity

()

Maximum Humidity

()

Mean Dew Point

(oC)

Minimum Dew Point

(oC)

Maximum Dew Point

(oC) Mean 2219 1447 3405 6223 2547 9516 1270 682 1844

Standard Error 036 034 041 076 057 068 029 032 029 Minimum 957 029 1752 2730 590 4830 -875 -1464 -098 Maximum 3460 2712 5066 9603 5400 10000 2366 1739 3613

Count 305 305 305 305 305 305 305 305 305 2008-2009

Mean 2561 1715 3884 7916 3825 10298 2022 1433 2736 Standard Error 031 030 035 058 069 018 028 029 029

Minimum 1110 331 2057 3455 1010 7040 532 -164 1103 Maximum 3644 2634 5855 10000 7870 10000 2985 2554 3877

Count 454 454 454 454 454 454 454 454 454

cycle (Fig 3) Air temperature monthly meansincreased during the summer (June throughSeptember) reaching as high as 322oC in August everyyear Air temperature minimum monthly meansdecreased during winter months (December throughFebruary) every year reaching as low as 130oC Airtemperature monthly and daily mean ranges fluctuatedwidely and ranged between 24 and 55oC and 30-41oCover the entire period of record yet followed the sameannual cycle (Fig 3) Monthly means ranged between30 to 32 and 35 to 41oC for winter and summer seasonrespectively

Air temperature monthly means for the nearbystations are based on reported daily means Airtemperature monthly means from all three stationsfollowed the same trends with air temperature peakingduring summer months reaching their lowest valuesduring winter months Monthly means from the studysite were higher and significantly different fromHurghadarsquos measurements Observed daily maximumsfrom the study site were also higher than Hurghadarsquosduring hot summer months yet very similar to IsmailiaDew point temperature followed the same annual cycleas air temperature decreasing as air temperature

1016

Moustafa MZ et al

Table 4 Results of spectral analysis for air temperature (07-08 08-09 and 07-09) measured at the study sitebased on half-hourly sampling intervals Regardless of the selected sampling intervals (ie half-hour vs hourly)

filtering the raw data (30-hours cut off frequency) vs no filter dominant cycles are yearly daily and half-daily

Frequency Period =

(1Frequency) days (2007_2008)

Amplitude = (Variance) 05 (2007_2008)

Period = (1Frequency) in days (2008_2009)

Amplitude = (Variance) 05 (2008_2009)

Period = (1Frequency)

days

Amplitude = (Variance)05

2007-2009 Mean 000 1970 000 2364 000 2337 Yearly 3413 148 34133 630 34133 605 Daily 100 393 100 462 100 489 Daily 100 218 100 223 100 176 Daily 100 142 100 139 100 131 Daily 099 073 100 123 100 088

Half-day 050 159 050 192 050 143 Half-day 050 098 050 119 050 135

Fig 4 Exceedence probability curves for observed daily minimum (Panel A) and maximum (Panel B) airtemperatures at Ain Sokhna (solid black) Hurghada (dark grey) and Ismailia (light grey) Minimum daily

temperature almost the same at study site and Ismailia located 200km north of study site and lower thanHurghada south of the study site by 400km by approximately 5oC Maximum air temperature at study site

exceeds Hurghadarsquos observed maximum temperature by an average of 7oC

decreased and increasing as air temperature increased(Fig 2) Dew point monthly means at the study sitewere significantly different for the entire period ofrecord with the exception of August of 2007 (due toshort record at the study site) and July 2007 (Fig 2)

Relative humidity monthly means were alsosignificantly different for most of the study periodstarting January of 2008 peaked during winter monthsdecreased during summer months and remained below100 throughout the two year period (Fig 2) Relative

1017

Int J Environ Res 9(3)1011-1022 Summer 2015

Fig 5 Schematic illustrating the hypothesized KatabaticFoehn winds at the study site This wind system isderived based on co-existence of a persistent year-round trade wind (north westerly wind) and Gebel Ataka lt

500m away from Zakirsquos Reef at Ain Sokhna

humidity annual cycle was opposite to air temperaturecycle increased when temperature decreased anddecreased when temperature increased Howeverobserved relative humidity values remained within thecomfort zone for humans (30-60 relative humidityrange) for more than 50 and 75 of the time duringthe first and second study years respectively (datanot shown)

Yearly air temperature means were 221 (plusmn007SE)and 256oC (plusmn008SE) during 2007-2008 and 2008-2009with an overall mean of 242oC (plusmn005SE) during 2007-2009 (Table 3) Observed differences between the twomeans is due to record length where number ofobservations for the first and second years was 14583and 21762 respectively However means calculatedbased on the same number of observations (N = 14583)from each year were almost identical at 221 (plusmn007SE)and 224 (plusmn007SE)

We used spectral analysis as the primary techniquefor assessing the cyclic nature of all univariate timeseries in the frequency domain At zero frequency (ietemperature spectrum oC2) the value is 504corresponding to mean square of the series (Table 4amplitude = 225oC) At the next frequency 0292969E-02 the value is 189 which approximately representsannual variations (15 and 63oC for 1st and 2nd yearrespectively) since 16384 data points at 30 minutescover 8192 hours (approximately 115 months)Remaining frequencies were dominated by daily cycleswith amplitudes ranging between 07 and 46oCfollowed by half-daily cycles with an amplitude of 10oC(Table 4) The strongest signals were yearly daily and

half-daily frequencies based on the resultantamplitudes Air temperature from the combined two-year data set revealed the same half day daily andyearly return periods all of which reveal that daily andhalf daily cycles are dominating the local climate withan amplitude of approximately 5oC Regardless of thedata set used (1st 2nd or the two-year combined) allspectral results are nearly identical showing thevariance at low frequency of yearly half-daily and dailyFurther all results are unaffected by sampling intervals(half-hourly hourly and de-seasonalized datasetsTable 4)

Frequency histograms revealed a bimodal signalwith one peak around 14-15oC and a second peakaround 32-33oC This bimodal signal was present infirst second and the two-year combined data setsThose two peaks represent the half-day cycle fromspectral analysis representing night (14-15oC) anddaytime (32-33oC) temperatures

The main reasons for the prolific development ofreef systems along the Red Sea proper are because ofits great depths (gt 1800 m) and efficient watercirculation pattern (Mustafa 2000) Very high surfacetemperatures coupled with high salinities make the RedSea one of the warmest and saltiest bodies of seawaterin the world The average surface water temperature ofthe Red Sea during the summer is about 26oC in thenorth and 30oC in the south Rainfall along the Red Seais extremely low averaging 006 m per year mostly inthe form of short spell showers There is virtually nosurface water runoff because no river enters the RedSea (Shahin 1989 Morcos 1970) The scarcity of

1018

Survival of Fringing Corals in Extreme Temperatures

rainfall and lack of a major source of fresh water resultin evaporation rates as high as 205 cmyr and highsalinity with minimal seasonal variation (Red Sea 2011)

It should be noted that of the approximately 35coral taxa known to survive in the Gulf of Suez onlysix (Acropora humilis A microclados A hemprichiiLitophyton arboretum Stylophora pistillata Poritescolumna and P plantulata (approximately 80 hardcorals and 20 soft corals) compose 94 of the reefrsquoscoral cover Coral dominance shifted over a four-yearperiod (2004-2007) mainly due to an oil spill in 2005global bleaching event (2005) and high (gt30oC) andlow (lt16oC) seawater temperature (Moustafa et al2008) Yet the six corals dominance remainedunchanged These corals regularly experience 4-65oCdaily changes in seawater temperature and seasonalvariations that exceed 29oC (Moustafa et al 2008)

There are no long-term studies in this region priorto the research reported here (Moustafa et al 2008a)This multi-year study set out to study the health ofZakirsquos Reef and its dominant coral species The factthat only six of the identified Gulf of Suez 35-40 coralspecies remained dominant for a five-year study(Moustafa et al 2008a) made us wonder what it wasabout these six species that allowed them to survive insuch a harsh environment Are these six species heattolerant or is there a linkage between fringing coralreefs located in the Gulf of Suez and the local climateInsight into local climate may provide a betterunderstanding of the direct influence weather has ifany on fringing coral reef health

What was found regarding air temperature at thestudy site and the two other locations was puzzling atfirst The fact that daily mean air temperature at thestudy site was less than the mean observed at Ismailiawas expected since the climatic trend in Egypt is thatair temperature increases going farther south Howeverit was not expected that air temperature daily means atthe study site and Hurghada would be very similar(Table 2) particularly if one considers how far Hurghadais from the study site (gt400 km south) Air temperaturedaily maximums at the study site exceeded bothHurghadarsquos and Ismailiarsquos by almost 7 and 9oCrespectively (Fig 4) While air temperature minimumdaily means are almost equal to Ismailiarsquos theminimums were less than Hurghadarsquos by 4 degreesThese observed trends were not expected consideringthe three stationsrsquo geographic locations suggestingthat other natural phenomena may be influencing airtemperature at the study site

Ain Sokhna Arabic for ldquohot springrdquo was namedafter the nearby sulfur springs These springs originateat Gebel Ataka (lt500 m West of Zakirsquos Reef) the

northern most mountains in the eastern desert Whenthe year-round trade wind approaches the study areait finds its first obstacle Gebel Ataka the mountainrange that starts at the study site and continues almostall the way south to Hurghada with an averageelevation of approximately 300m Gebel Ataka was thefirst factor that we believed may be influencing thelocal climate We hypothesized that these highplateauing mountains force the trade winds upwardand as a result it loses its moisture content anddecreases in temperature Air parcels reaching thehighest point on top of Gebel Ataka start descendingdue to gravity (dense air) on the leeside (down-wind)and towards the study site This wind pattern at thestudy site mimics a Katabatic wind system (Fig 5)

Katabatic wind is the technical name for a windthat carries high density air from a higher elevationdown a slope under the force of gravity Katabaticwinds originate from radiational cooling of air atop aplateau a mountain or even a hill Since the density ofair is inversely proportional to temperature the air willflow downwards warming adiabatically as it descendsThe temperature of the wind depends on thetemperature in the source region and height of descentAt Ain Sokhna it is possible that due to the presenceof this local mountain an adiabatic wind pattern existslocally and causes the air temperature to rise at thestudy site as becomes evident when one comparesthe study site air temperature observations with thosefrom Hurghada

The entire near-surface wind field and airtemperature at sea level at the study site may be heavilyinfluenced by the Katabatic winds What differentiatesthe study site from the other two stations is the factthat the mountain range is less than 500m from thestudy site and no major obstacles exist between AtakaMountain and Zakirsquos Reef (Fig 1b) The close proximityof the study site to the mountain means that elevatedtemperatures air parcels descending from Gebel Atakaare likely to cause a marked increase of air temperatureat the study site In contrast one of the main reasonsthat Hurghadarsquos temperatures are less than thoseobserved at the study site is because Ataka mountainrange is located very far (gt30 km) from the Red Seacoast For example mean observations at the studysite are very close to temperature means at Hurghada(242 vs 252oC Table 2) yet daily maximumobservations at the study site exceed Hurghada andIsmailia by almost 7 and 9 degrees respectively (Fig4) Furthermore during winter time the existence ofthe warm dry Katabatic (Foehn) wind increases airtemperature at the study site and consequently causesthe minimum daily means at the study site and Ismailiato be almost equal (Fig 4) Those warm air parcels are

1019

Int J Environ Res 9(3)1011-1022 Summer 2015

Fig 6 NOAArsquos HySplit Model results Panel A depicts forward trajectory which represents airmovement over a one week time span originating at a point approximately 10km away from ZakirsquosReef (behind the western mountain) and shows that the winds at the reefrsquos vicinity originated from

behind the western mountain all of which confirm the ascending and descending wind patternrequired for a Foehn wind Panel B depicts backward trajectory of air movement where Zakirsquos Reeflocation is the reference point This image highlights the distance traveled from the windrsquos point oforigin to the chosen endpoint (Zakirsquos Reef) and highlights the presence of a dominant westerly windin the region Both forward and backward trajectories depicted further support the existence of a

Foehn wind near the reef as a result of the consistent westerly wind and the surmounting of thepreceding mountain range

1020

Moustafa MZ et al

likely to raise water temperature particularly duringwinter season and help resident corals to survive atthis location

Warm dry Katabatic winds occur on the leeside ofa mountain range These Katabatic winds that descendon the leeside of a mountain (eg Gebel Ataka) andwarmed adiabatically are called Foehn winds (alsoChinook or Bergwind Fig 5) As a consequence of thedifferent adiabatic lapse rates of moist and dry air theair on the leeward slopes becomes warmer thanequivalent elevations on the windward slopes and canraise temperatures by as much as 30degC in just a matterof hours (Hess and McKnight 2000) For example oneof the main reasons that central Europe enjoys a warmerclimate is due to the Foehn as moisture-filled windsoff the Mediterranean Sea blow over the Alps In ourcase the end result is that conditions on the upwindside of Gebel Ataka will be relatively moist and coolwhile conditions on the leeward side of the mountaincorresponding to the study site will be relatively warmand dry As the air flow descends on the leeward sideof Gebel Ataka it becomes compressed by atmosphericpressure which raises the air-flow temperature Inaddition to compression the air flow also gains moreheat directly from the sun (particularly during summermonths) when the skies are typically clear

The increase in air temperature at sea level mayhave some positive consequences for corals at thislocation Warmer than normal air temperature due tothis local wind pattern is more likely to increase seawatertemperature at Zakirsquos Reef Comparisons between airand seawater temperature collected at Zakirsquos Reef revealthat their means are almost identical (Moustafa et al2008b 2013 2014) We speculate that the two yearlymeans of air and seawater temperatures are almostidentical as a result of the local wind pattern This warmmountain air elevates seawater temperature all of whichmay help explain the existence and survival of ZakirsquosReef at such a high latitude

HYSPLIT trajectory model was used to serve asan independent validation for the existence of thenewly identified Foehn wind Model runs ranged fromdaily to weekly traces and quantified average flow rateof the air masses and temperature rates of change asthey encountered resistances caused by naturalobstructions (eg mountain ranges) After initializingmodel runs model parameters were adjusted (egnumber of starting points temporal scale of trajectoriesand frequency of extraction) in order to observeresulting model outputs These extractions wereprojected and utilized in Google Earth to serve as avisual aid of model run outputs The resulting HYSPLITmodel data served as independent validation to supportthe proposed hypothesis regarding the existence of

Foehn winds in the Red Sea region near the study site(Fig 6)

Two HYSPLIT model run results are displayed inFig 6 The first trajectory (Fig 6a) is a forward trajectoryof air movement over a one week period originating ata point approximately 10 km West of Gabel AtakaMountain (29o 22rsquo 3047 N and 32o 15rsquo 1025 E) Thismodel run predicts that the winds in Zakirsquos Reefvicinity originate from a point behind the westwardmountain range likely confirming the risingfalling windpattern - a signature of a Foehn wind The backwardtrajectory in Fig 6b illustrates the distance traveledfrom the windrsquos point of origin to the predefined endpoint (Zakirsquos Reef) and depicts the presence of adominant westerly wind in the region These twotrajectories depicted in Fig 6 further support theexistence of a Foehn wind near Zakirsquos reef as a resultof the consistent westerly wind and the surmountingof the Ataka mountain rangeSeawater temperature is akey indicator of coral reef health Seawater temperatureinfluences ecological processes and is considered theprimary factor limiting coral reef survival anddevelopment throughout the world (Szmant ampGassman 1990 Goreau amp Hayes 2005) Seawatertemperature affects coral breeding spawning larvalsettlement and growth (Szmant amp Gassman 1990Mendes amp Woody 2002) and coral recruitment (Jokielamp Coles 1990)

It is evident that Zakirsquos coral species have arelatively higher thermal tolerance compared to similarspecies at other locations only the most hot and cold-tolerant few have been identified at this site (Moustafaet al 2008a 2013) The world-wide bleaching event in2005 resulted in lethal heat conditions to coral speciesworldwide yet coral species found at Zakirsquos reef survivetemperatures more extreme than most every yearRegardless of all exposures to lethal or near-lethal limitsof hot and cold the six coral species at Zakirsquos reefremained dominant (gt94)

Lirman et al (2011) suggested that the only coraltaxa that survive the cold event with limited mortalitywere those that are also found in other habitats thatoften experience low temperatures on a regular basissuggesting species-specific adaptive resistancemechanisms Corals on Zakirsquos reef experience theseextreme seawater temperatures annually as reportedby Moustafa et al (2014) for both summer and winterExposure to these high- and low-temperature anomalieson a yearly basis may have caused corals at Zakirsquos reefto develop an adaptive resistance mechanism It is nosurprise that only six out of the 35 identified coralspecies in the Gulf of Suez dominate Zakirsquos reef (gt94)and can be attributed to their adaptive resistancemechanisms to both low and high extreme temperatures

1021

Int J Environ Res 9(3)1011-1022 Summer 2015

These corals have been living at the study site forhundreds of years and may have the capacity toacclimatize or adapt to external stresses better thanothers For example previous comparisons amongplants (Ingram amp Bartels 1996) and animals (Huey ampKingsolver 1993 Somero 2002 Hochachka amp Somero2002) have shown that some population or species arefar more resilient compared to others with markedinfluences on growth survival and disease resistance(Nevo 2001 Hochachka amp Somero 2002 Parsons2005) Furthermore Barshis et al (2012) studied coralgenes and concluded that some of these genes maybestow resilience through faster reaction duringtransient heat stress which may prepare corals forfrequently encountered stress

Coles and Fadlallah (1991) reported that Acroporapharaonis and Porites compressa in the Arabian Gulfsuffered severe mortality when the water temperaturefell below 115oC Furthermore they reported thatPorites compressa (the dominant species at their reef)and Faviidae corals showed sign of sub-lethal effectswhen temperature was below 12oC for two consecutivedays yet recovered after six months Some of the coralgenera found in the Arabian Gulf are also found inZakirsquos reef (eg Acropora humilis and Poritescolumna) Unlike the Arabian Gulf counterpartresident Acropora in Zakirsquos reef survived well duringthe worldwide bleaching event in 2005 as well asduring cold temperatures measured in the 2006 and2007 winter seasons (Moustafa et al 2014)

CONCLUSIONSThe current study was driven mainly by scientific

curiosity regarding the discovery of a fringing coralreef ecosystem that thrived at one of the highestlatitudes of any sub-tropical reef system in the worldWe quickly learned that there are no significantenvironmental databases for this region except for thevery different Gulf of Aqaba We therefore integratedother environmental data to help us find out why thisecosystem is flourishing at such a high latitude

Our data collection and analysis lead to severalnote-worthy discoveries most importantly theexistence of a KatabaticFoehn wind system causedby the co-existence of the high mountain range andthe dominant trade winds at the study site This windsystem causes elevated temperatures during the hotsummer and winter months at the study site Thislocation is the only one where the mountains are veryclose to the coast along the entire Egyptian Red SeaAs air temperature increases water temperature at thestudy site may also increase Consequentlytemperature rise particularly during winter seasonsenhances Zakirsquos Reefrsquos chances of survival at such

high latitude Furthermore this warm local wind mayhave also helped the corals at this location evolveacclimate to the extreme conditions prevalent duringsummer months The entire Gulf of Suez fringing coralreef ecosystem may be reliant for survival particularlyduring winter months on this newly discovered localwind pattern

ACKNOWLEDGEMENTSSpecial thanks to all family members particularly

my son Zachariah Moustafa for providing fieldassistance and being the cause for expanding ourresearch In 2007 Zachariahrsquos expression caught in aphotograph from the incredible summer heat at thestudy site (registered at 135 Fahrenheit) caused us tore-evaluate our plans and start collecting meteorologicaldata Catherine Miller from Dreyfoos School of the Artsprovided lab equipment and helped find funding It iswith great appreciations to PADI and North Carolinaresearch consortium for their financial support of thisresearch Without the support of Onset and YSICorporation graciously donating and loaning fieldequipment the culmination of this research would notbeen possible particularly the dedicated staff especiallyKory Wagner from Onset and their technical supportthroughout this research project

REFERENCES

Stein A F Isakov V Godowitch J and Draxler R R(2007) A hybrid modeling approach to resolve pollutantconcentrations in an urban area Atmospheric Environment41 (40) 9410-9428

Barshis DJ Lander JT Oliver TA Seneca FO Taylor-knowles N and Palumbi SR (2012) Genomic basis forcoral resilience to climate change Proceedings of the NationalAcademy of Sciences of the United States of America 110(4) 1387-1392

Coles SL and Fadlallah YH (1991) Reef coral survivaland mortality at low temperatures in the Arabian Gulf Newspecies-specific lower temperature limits Coral Reefs 9(4)231-237

Draxler R R and Rolph G D (2013) HYSPLIT (HYbridSingle-Particle Lagrangian Integrated Trajectory) Modelaccess via NOAA ARL READY Website httpreadyarlnoaagovHYSPLITphp NOAA Air ResourcesLaboratory Silver Spring MD

Jokiel P L and Coles S L (1990) Response of Hawaiianand other Indo-Pacific reef corals to elevated temperatureCoral Reefs 8 155-162

Goreau TJF and Hayes RL (2005) Global coral reefbleaching and sea surface temperature trends from satellite-derived hotspot analysis World Resource Review 17(2) 254-293

Hanaur E 1(988) The Egyptian Red Sea A Diverrsquos Guide(San Diego CA Watersport Publishing Inc)

1016

Moustafa MZ et al

Table 4 Results of spectral analysis for air temperature (07-08 08-09 and 07-09) measured at the study sitebased on half-hourly sampling intervals Regardless of the selected sampling intervals (ie half-hour vs hourly)

filtering the raw data (30-hours cut off frequency) vs no filter dominant cycles are yearly daily and half-daily

Frequency Period =

(1Frequency) days (2007_2008)

Amplitude = (Variance) 05 (2007_2008)

Period = (1Frequency) in days (2008_2009)

Amplitude = (Variance) 05 (2008_2009)

Period = (1Frequency)

days

Amplitude = (Variance)05

2007-2009 Mean 000 1970 000 2364 000 2337 Yearly 3413 148 34133 630 34133 605 Daily 100 393 100 462 100 489 Daily 100 218 100 223 100 176 Daily 100 142 100 139 100 131 Daily 099 073 100 123 100 088

Half-day 050 159 050 192 050 143 Half-day 050 098 050 119 050 135

Fig 4 Exceedence probability curves for observed daily minimum (Panel A) and maximum (Panel B) airtemperatures at Ain Sokhna (solid black) Hurghada (dark grey) and Ismailia (light grey) Minimum daily

temperature almost the same at study site and Ismailia located 200km north of study site and lower thanHurghada south of the study site by 400km by approximately 5oC Maximum air temperature at study site

exceeds Hurghadarsquos observed maximum temperature by an average of 7oC

decreased and increasing as air temperature increased(Fig 2) Dew point monthly means at the study sitewere significantly different for the entire period ofrecord with the exception of August of 2007 (due toshort record at the study site) and July 2007 (Fig 2)

Relative humidity monthly means were alsosignificantly different for most of the study periodstarting January of 2008 peaked during winter monthsdecreased during summer months and remained below100 throughout the two year period (Fig 2) Relative

1017

Int J Environ Res 9(3)1011-1022 Summer 2015

Fig 5 Schematic illustrating the hypothesized KatabaticFoehn winds at the study site This wind system isderived based on co-existence of a persistent year-round trade wind (north westerly wind) and Gebel Ataka lt

500m away from Zakirsquos Reef at Ain Sokhna

humidity annual cycle was opposite to air temperaturecycle increased when temperature decreased anddecreased when temperature increased Howeverobserved relative humidity values remained within thecomfort zone for humans (30-60 relative humidityrange) for more than 50 and 75 of the time duringthe first and second study years respectively (datanot shown)

Yearly air temperature means were 221 (plusmn007SE)and 256oC (plusmn008SE) during 2007-2008 and 2008-2009with an overall mean of 242oC (plusmn005SE) during 2007-2009 (Table 3) Observed differences between the twomeans is due to record length where number ofobservations for the first and second years was 14583and 21762 respectively However means calculatedbased on the same number of observations (N = 14583)from each year were almost identical at 221 (plusmn007SE)and 224 (plusmn007SE)

We used spectral analysis as the primary techniquefor assessing the cyclic nature of all univariate timeseries in the frequency domain At zero frequency (ietemperature spectrum oC2) the value is 504corresponding to mean square of the series (Table 4amplitude = 225oC) At the next frequency 0292969E-02 the value is 189 which approximately representsannual variations (15 and 63oC for 1st and 2nd yearrespectively) since 16384 data points at 30 minutescover 8192 hours (approximately 115 months)Remaining frequencies were dominated by daily cycleswith amplitudes ranging between 07 and 46oCfollowed by half-daily cycles with an amplitude of 10oC(Table 4) The strongest signals were yearly daily and

half-daily frequencies based on the resultantamplitudes Air temperature from the combined two-year data set revealed the same half day daily andyearly return periods all of which reveal that daily andhalf daily cycles are dominating the local climate withan amplitude of approximately 5oC Regardless of thedata set used (1st 2nd or the two-year combined) allspectral results are nearly identical showing thevariance at low frequency of yearly half-daily and dailyFurther all results are unaffected by sampling intervals(half-hourly hourly and de-seasonalized datasetsTable 4)

Frequency histograms revealed a bimodal signalwith one peak around 14-15oC and a second peakaround 32-33oC This bimodal signal was present infirst second and the two-year combined data setsThose two peaks represent the half-day cycle fromspectral analysis representing night (14-15oC) anddaytime (32-33oC) temperatures

The main reasons for the prolific development ofreef systems along the Red Sea proper are because ofits great depths (gt 1800 m) and efficient watercirculation pattern (Mustafa 2000) Very high surfacetemperatures coupled with high salinities make the RedSea one of the warmest and saltiest bodies of seawaterin the world The average surface water temperature ofthe Red Sea during the summer is about 26oC in thenorth and 30oC in the south Rainfall along the Red Seais extremely low averaging 006 m per year mostly inthe form of short spell showers There is virtually nosurface water runoff because no river enters the RedSea (Shahin 1989 Morcos 1970) The scarcity of

1018

Survival of Fringing Corals in Extreme Temperatures

rainfall and lack of a major source of fresh water resultin evaporation rates as high as 205 cmyr and highsalinity with minimal seasonal variation (Red Sea 2011)

It should be noted that of the approximately 35coral taxa known to survive in the Gulf of Suez onlysix (Acropora humilis A microclados A hemprichiiLitophyton arboretum Stylophora pistillata Poritescolumna and P plantulata (approximately 80 hardcorals and 20 soft corals) compose 94 of the reefrsquoscoral cover Coral dominance shifted over a four-yearperiod (2004-2007) mainly due to an oil spill in 2005global bleaching event (2005) and high (gt30oC) andlow (lt16oC) seawater temperature (Moustafa et al2008) Yet the six corals dominance remainedunchanged These corals regularly experience 4-65oCdaily changes in seawater temperature and seasonalvariations that exceed 29oC (Moustafa et al 2008)

There are no long-term studies in this region priorto the research reported here (Moustafa et al 2008a)This multi-year study set out to study the health ofZakirsquos Reef and its dominant coral species The factthat only six of the identified Gulf of Suez 35-40 coralspecies remained dominant for a five-year study(Moustafa et al 2008a) made us wonder what it wasabout these six species that allowed them to survive insuch a harsh environment Are these six species heattolerant or is there a linkage between fringing coralreefs located in the Gulf of Suez and the local climateInsight into local climate may provide a betterunderstanding of the direct influence weather has ifany on fringing coral reef health

What was found regarding air temperature at thestudy site and the two other locations was puzzling atfirst The fact that daily mean air temperature at thestudy site was less than the mean observed at Ismailiawas expected since the climatic trend in Egypt is thatair temperature increases going farther south Howeverit was not expected that air temperature daily means atthe study site and Hurghada would be very similar(Table 2) particularly if one considers how far Hurghadais from the study site (gt400 km south) Air temperaturedaily maximums at the study site exceeded bothHurghadarsquos and Ismailiarsquos by almost 7 and 9oCrespectively (Fig 4) While air temperature minimumdaily means are almost equal to Ismailiarsquos theminimums were less than Hurghadarsquos by 4 degreesThese observed trends were not expected consideringthe three stationsrsquo geographic locations suggestingthat other natural phenomena may be influencing airtemperature at the study site

Ain Sokhna Arabic for ldquohot springrdquo was namedafter the nearby sulfur springs These springs originateat Gebel Ataka (lt500 m West of Zakirsquos Reef) the

northern most mountains in the eastern desert Whenthe year-round trade wind approaches the study areait finds its first obstacle Gebel Ataka the mountainrange that starts at the study site and continues almostall the way south to Hurghada with an averageelevation of approximately 300m Gebel Ataka was thefirst factor that we believed may be influencing thelocal climate We hypothesized that these highplateauing mountains force the trade winds upwardand as a result it loses its moisture content anddecreases in temperature Air parcels reaching thehighest point on top of Gebel Ataka start descendingdue to gravity (dense air) on the leeside (down-wind)and towards the study site This wind pattern at thestudy site mimics a Katabatic wind system (Fig 5)

Katabatic wind is the technical name for a windthat carries high density air from a higher elevationdown a slope under the force of gravity Katabaticwinds originate from radiational cooling of air atop aplateau a mountain or even a hill Since the density ofair is inversely proportional to temperature the air willflow downwards warming adiabatically as it descendsThe temperature of the wind depends on thetemperature in the source region and height of descentAt Ain Sokhna it is possible that due to the presenceof this local mountain an adiabatic wind pattern existslocally and causes the air temperature to rise at thestudy site as becomes evident when one comparesthe study site air temperature observations with thosefrom Hurghada

The entire near-surface wind field and airtemperature at sea level at the study site may be heavilyinfluenced by the Katabatic winds What differentiatesthe study site from the other two stations is the factthat the mountain range is less than 500m from thestudy site and no major obstacles exist between AtakaMountain and Zakirsquos Reef (Fig 1b) The close proximityof the study site to the mountain means that elevatedtemperatures air parcels descending from Gebel Atakaare likely to cause a marked increase of air temperatureat the study site In contrast one of the main reasonsthat Hurghadarsquos temperatures are less than thoseobserved at the study site is because Ataka mountainrange is located very far (gt30 km) from the Red Seacoast For example mean observations at the studysite are very close to temperature means at Hurghada(242 vs 252oC Table 2) yet daily maximumobservations at the study site exceed Hurghada andIsmailia by almost 7 and 9 degrees respectively (Fig4) Furthermore during winter time the existence ofthe warm dry Katabatic (Foehn) wind increases airtemperature at the study site and consequently causesthe minimum daily means at the study site and Ismailiato be almost equal (Fig 4) Those warm air parcels are

1019

Int J Environ Res 9(3)1011-1022 Summer 2015

Fig 6 NOAArsquos HySplit Model results Panel A depicts forward trajectory which represents airmovement over a one week time span originating at a point approximately 10km away from ZakirsquosReef (behind the western mountain) and shows that the winds at the reefrsquos vicinity originated from

behind the western mountain all of which confirm the ascending and descending wind patternrequired for a Foehn wind Panel B depicts backward trajectory of air movement where Zakirsquos Reeflocation is the reference point This image highlights the distance traveled from the windrsquos point oforigin to the chosen endpoint (Zakirsquos Reef) and highlights the presence of a dominant westerly windin the region Both forward and backward trajectories depicted further support the existence of a

Foehn wind near the reef as a result of the consistent westerly wind and the surmounting of thepreceding mountain range

1020

Moustafa MZ et al

likely to raise water temperature particularly duringwinter season and help resident corals to survive atthis location

Warm dry Katabatic winds occur on the leeside ofa mountain range These Katabatic winds that descendon the leeside of a mountain (eg Gebel Ataka) andwarmed adiabatically are called Foehn winds (alsoChinook or Bergwind Fig 5) As a consequence of thedifferent adiabatic lapse rates of moist and dry air theair on the leeward slopes becomes warmer thanequivalent elevations on the windward slopes and canraise temperatures by as much as 30degC in just a matterof hours (Hess and McKnight 2000) For example oneof the main reasons that central Europe enjoys a warmerclimate is due to the Foehn as moisture-filled windsoff the Mediterranean Sea blow over the Alps In ourcase the end result is that conditions on the upwindside of Gebel Ataka will be relatively moist and coolwhile conditions on the leeward side of the mountaincorresponding to the study site will be relatively warmand dry As the air flow descends on the leeward sideof Gebel Ataka it becomes compressed by atmosphericpressure which raises the air-flow temperature Inaddition to compression the air flow also gains moreheat directly from the sun (particularly during summermonths) when the skies are typically clear

The increase in air temperature at sea level mayhave some positive consequences for corals at thislocation Warmer than normal air temperature due tothis local wind pattern is more likely to increase seawatertemperature at Zakirsquos Reef Comparisons between airand seawater temperature collected at Zakirsquos Reef revealthat their means are almost identical (Moustafa et al2008b 2013 2014) We speculate that the two yearlymeans of air and seawater temperatures are almostidentical as a result of the local wind pattern This warmmountain air elevates seawater temperature all of whichmay help explain the existence and survival of ZakirsquosReef at such a high latitude

HYSPLIT trajectory model was used to serve asan independent validation for the existence of thenewly identified Foehn wind Model runs ranged fromdaily to weekly traces and quantified average flow rateof the air masses and temperature rates of change asthey encountered resistances caused by naturalobstructions (eg mountain ranges) After initializingmodel runs model parameters were adjusted (egnumber of starting points temporal scale of trajectoriesand frequency of extraction) in order to observeresulting model outputs These extractions wereprojected and utilized in Google Earth to serve as avisual aid of model run outputs The resulting HYSPLITmodel data served as independent validation to supportthe proposed hypothesis regarding the existence of

Foehn winds in the Red Sea region near the study site(Fig 6)

Two HYSPLIT model run results are displayed inFig 6 The first trajectory (Fig 6a) is a forward trajectoryof air movement over a one week period originating ata point approximately 10 km West of Gabel AtakaMountain (29o 22rsquo 3047 N and 32o 15rsquo 1025 E) Thismodel run predicts that the winds in Zakirsquos Reefvicinity originate from a point behind the westwardmountain range likely confirming the risingfalling windpattern - a signature of a Foehn wind The backwardtrajectory in Fig 6b illustrates the distance traveledfrom the windrsquos point of origin to the predefined endpoint (Zakirsquos Reef) and depicts the presence of adominant westerly wind in the region These twotrajectories depicted in Fig 6 further support theexistence of a Foehn wind near Zakirsquos reef as a resultof the consistent westerly wind and the surmountingof the Ataka mountain rangeSeawater temperature is akey indicator of coral reef health Seawater temperatureinfluences ecological processes and is considered theprimary factor limiting coral reef survival anddevelopment throughout the world (Szmant ampGassman 1990 Goreau amp Hayes 2005) Seawatertemperature affects coral breeding spawning larvalsettlement and growth (Szmant amp Gassman 1990Mendes amp Woody 2002) and coral recruitment (Jokielamp Coles 1990)

It is evident that Zakirsquos coral species have arelatively higher thermal tolerance compared to similarspecies at other locations only the most hot and cold-tolerant few have been identified at this site (Moustafaet al 2008a 2013) The world-wide bleaching event in2005 resulted in lethal heat conditions to coral speciesworldwide yet coral species found at Zakirsquos reef survivetemperatures more extreme than most every yearRegardless of all exposures to lethal or near-lethal limitsof hot and cold the six coral species at Zakirsquos reefremained dominant (gt94)

Lirman et al (2011) suggested that the only coraltaxa that survive the cold event with limited mortalitywere those that are also found in other habitats thatoften experience low temperatures on a regular basissuggesting species-specific adaptive resistancemechanisms Corals on Zakirsquos reef experience theseextreme seawater temperatures annually as reportedby Moustafa et al (2014) for both summer and winterExposure to these high- and low-temperature anomalieson a yearly basis may have caused corals at Zakirsquos reefto develop an adaptive resistance mechanism It is nosurprise that only six out of the 35 identified coralspecies in the Gulf of Suez dominate Zakirsquos reef (gt94)and can be attributed to their adaptive resistancemechanisms to both low and high extreme temperatures

1021

Int J Environ Res 9(3)1011-1022 Summer 2015

These corals have been living at the study site forhundreds of years and may have the capacity toacclimatize or adapt to external stresses better thanothers For example previous comparisons amongplants (Ingram amp Bartels 1996) and animals (Huey ampKingsolver 1993 Somero 2002 Hochachka amp Somero2002) have shown that some population or species arefar more resilient compared to others with markedinfluences on growth survival and disease resistance(Nevo 2001 Hochachka amp Somero 2002 Parsons2005) Furthermore Barshis et al (2012) studied coralgenes and concluded that some of these genes maybestow resilience through faster reaction duringtransient heat stress which may prepare corals forfrequently encountered stress

Coles and Fadlallah (1991) reported that Acroporapharaonis and Porites compressa in the Arabian Gulfsuffered severe mortality when the water temperaturefell below 115oC Furthermore they reported thatPorites compressa (the dominant species at their reef)and Faviidae corals showed sign of sub-lethal effectswhen temperature was below 12oC for two consecutivedays yet recovered after six months Some of the coralgenera found in the Arabian Gulf are also found inZakirsquos reef (eg Acropora humilis and Poritescolumna) Unlike the Arabian Gulf counterpartresident Acropora in Zakirsquos reef survived well duringthe worldwide bleaching event in 2005 as well asduring cold temperatures measured in the 2006 and2007 winter seasons (Moustafa et al 2014)

CONCLUSIONSThe current study was driven mainly by scientific

curiosity regarding the discovery of a fringing coralreef ecosystem that thrived at one of the highestlatitudes of any sub-tropical reef system in the worldWe quickly learned that there are no significantenvironmental databases for this region except for thevery different Gulf of Aqaba We therefore integratedother environmental data to help us find out why thisecosystem is flourishing at such a high latitude

Our data collection and analysis lead to severalnote-worthy discoveries most importantly theexistence of a KatabaticFoehn wind system causedby the co-existence of the high mountain range andthe dominant trade winds at the study site This windsystem causes elevated temperatures during the hotsummer and winter months at the study site Thislocation is the only one where the mountains are veryclose to the coast along the entire Egyptian Red SeaAs air temperature increases water temperature at thestudy site may also increase Consequentlytemperature rise particularly during winter seasonsenhances Zakirsquos Reefrsquos chances of survival at such

high latitude Furthermore this warm local wind mayhave also helped the corals at this location evolveacclimate to the extreme conditions prevalent duringsummer months The entire Gulf of Suez fringing coralreef ecosystem may be reliant for survival particularlyduring winter months on this newly discovered localwind pattern

ACKNOWLEDGEMENTSSpecial thanks to all family members particularly

my son Zachariah Moustafa for providing fieldassistance and being the cause for expanding ourresearch In 2007 Zachariahrsquos expression caught in aphotograph from the incredible summer heat at thestudy site (registered at 135 Fahrenheit) caused us tore-evaluate our plans and start collecting meteorologicaldata Catherine Miller from Dreyfoos School of the Artsprovided lab equipment and helped find funding It iswith great appreciations to PADI and North Carolinaresearch consortium for their financial support of thisresearch Without the support of Onset and YSICorporation graciously donating and loaning fieldequipment the culmination of this research would notbeen possible particularly the dedicated staff especiallyKory Wagner from Onset and their technical supportthroughout this research project

REFERENCES

Stein A F Isakov V Godowitch J and Draxler R R(2007) A hybrid modeling approach to resolve pollutantconcentrations in an urban area Atmospheric Environment41 (40) 9410-9428

Barshis DJ Lander JT Oliver TA Seneca FO Taylor-knowles N and Palumbi SR (2012) Genomic basis forcoral resilience to climate change Proceedings of the NationalAcademy of Sciences of the United States of America 110(4) 1387-1392

Coles SL and Fadlallah YH (1991) Reef coral survivaland mortality at low temperatures in the Arabian Gulf Newspecies-specific lower temperature limits Coral Reefs 9(4)231-237

Draxler R R and Rolph G D (2013) HYSPLIT (HYbridSingle-Particle Lagrangian Integrated Trajectory) Modelaccess via NOAA ARL READY Website httpreadyarlnoaagovHYSPLITphp NOAA Air ResourcesLaboratory Silver Spring MD

Jokiel P L and Coles S L (1990) Response of Hawaiianand other Indo-Pacific reef corals to elevated temperatureCoral Reefs 8 155-162

Goreau TJF and Hayes RL (2005) Global coral reefbleaching and sea surface temperature trends from satellite-derived hotspot analysis World Resource Review 17(2) 254-293

Hanaur E 1(988) The Egyptian Red Sea A Diverrsquos Guide(San Diego CA Watersport Publishing Inc)

1017

Int J Environ Res 9(3)1011-1022 Summer 2015

Fig 5 Schematic illustrating the hypothesized KatabaticFoehn winds at the study site This wind system isderived based on co-existence of a persistent year-round trade wind (north westerly wind) and Gebel Ataka lt

500m away from Zakirsquos Reef at Ain Sokhna

humidity annual cycle was opposite to air temperaturecycle increased when temperature decreased anddecreased when temperature increased Howeverobserved relative humidity values remained within thecomfort zone for humans (30-60 relative humidityrange) for more than 50 and 75 of the time duringthe first and second study years respectively (datanot shown)

Yearly air temperature means were 221 (plusmn007SE)and 256oC (plusmn008SE) during 2007-2008 and 2008-2009with an overall mean of 242oC (plusmn005SE) during 2007-2009 (Table 3) Observed differences between the twomeans is due to record length where number ofobservations for the first and second years was 14583and 21762 respectively However means calculatedbased on the same number of observations (N = 14583)from each year were almost identical at 221 (plusmn007SE)and 224 (plusmn007SE)

We used spectral analysis as the primary techniquefor assessing the cyclic nature of all univariate timeseries in the frequency domain At zero frequency (ietemperature spectrum oC2) the value is 504corresponding to mean square of the series (Table 4amplitude = 225oC) At the next frequency 0292969E-02 the value is 189 which approximately representsannual variations (15 and 63oC for 1st and 2nd yearrespectively) since 16384 data points at 30 minutescover 8192 hours (approximately 115 months)Remaining frequencies were dominated by daily cycleswith amplitudes ranging between 07 and 46oCfollowed by half-daily cycles with an amplitude of 10oC(Table 4) The strongest signals were yearly daily and

half-daily frequencies based on the resultantamplitudes Air temperature from the combined two-year data set revealed the same half day daily andyearly return periods all of which reveal that daily andhalf daily cycles are dominating the local climate withan amplitude of approximately 5oC Regardless of thedata set used (1st 2nd or the two-year combined) allspectral results are nearly identical showing thevariance at low frequency of yearly half-daily and dailyFurther all results are unaffected by sampling intervals(half-hourly hourly and de-seasonalized datasetsTable 4)

Frequency histograms revealed a bimodal signalwith one peak around 14-15oC and a second peakaround 32-33oC This bimodal signal was present infirst second and the two-year combined data setsThose two peaks represent the half-day cycle fromspectral analysis representing night (14-15oC) anddaytime (32-33oC) temperatures

The main reasons for the prolific development ofreef systems along the Red Sea proper are because ofits great depths (gt 1800 m) and efficient watercirculation pattern (Mustafa 2000) Very high surfacetemperatures coupled with high salinities make the RedSea one of the warmest and saltiest bodies of seawaterin the world The average surface water temperature ofthe Red Sea during the summer is about 26oC in thenorth and 30oC in the south Rainfall along the Red Seais extremely low averaging 006 m per year mostly inthe form of short spell showers There is virtually nosurface water runoff because no river enters the RedSea (Shahin 1989 Morcos 1970) The scarcity of

1018

Survival of Fringing Corals in Extreme Temperatures

rainfall and lack of a major source of fresh water resultin evaporation rates as high as 205 cmyr and highsalinity with minimal seasonal variation (Red Sea 2011)

It should be noted that of the approximately 35coral taxa known to survive in the Gulf of Suez onlysix (Acropora humilis A microclados A hemprichiiLitophyton arboretum Stylophora pistillata Poritescolumna and P plantulata (approximately 80 hardcorals and 20 soft corals) compose 94 of the reefrsquoscoral cover Coral dominance shifted over a four-yearperiod (2004-2007) mainly due to an oil spill in 2005global bleaching event (2005) and high (gt30oC) andlow (lt16oC) seawater temperature (Moustafa et al2008) Yet the six corals dominance remainedunchanged These corals regularly experience 4-65oCdaily changes in seawater temperature and seasonalvariations that exceed 29oC (Moustafa et al 2008)

There are no long-term studies in this region priorto the research reported here (Moustafa et al 2008a)This multi-year study set out to study the health ofZakirsquos Reef and its dominant coral species The factthat only six of the identified Gulf of Suez 35-40 coralspecies remained dominant for a five-year study(Moustafa et al 2008a) made us wonder what it wasabout these six species that allowed them to survive insuch a harsh environment Are these six species heattolerant or is there a linkage between fringing coralreefs located in the Gulf of Suez and the local climateInsight into local climate may provide a betterunderstanding of the direct influence weather has ifany on fringing coral reef health

What was found regarding air temperature at thestudy site and the two other locations was puzzling atfirst The fact that daily mean air temperature at thestudy site was less than the mean observed at Ismailiawas expected since the climatic trend in Egypt is thatair temperature increases going farther south Howeverit was not expected that air temperature daily means atthe study site and Hurghada would be very similar(Table 2) particularly if one considers how far Hurghadais from the study site (gt400 km south) Air temperaturedaily maximums at the study site exceeded bothHurghadarsquos and Ismailiarsquos by almost 7 and 9oCrespectively (Fig 4) While air temperature minimumdaily means are almost equal to Ismailiarsquos theminimums were less than Hurghadarsquos by 4 degreesThese observed trends were not expected consideringthe three stationsrsquo geographic locations suggestingthat other natural phenomena may be influencing airtemperature at the study site

Ain Sokhna Arabic for ldquohot springrdquo was namedafter the nearby sulfur springs These springs originateat Gebel Ataka (lt500 m West of Zakirsquos Reef) the

northern most mountains in the eastern desert Whenthe year-round trade wind approaches the study areait finds its first obstacle Gebel Ataka the mountainrange that starts at the study site and continues almostall the way south to Hurghada with an averageelevation of approximately 300m Gebel Ataka was thefirst factor that we believed may be influencing thelocal climate We hypothesized that these highplateauing mountains force the trade winds upwardand as a result it loses its moisture content anddecreases in temperature Air parcels reaching thehighest point on top of Gebel Ataka start descendingdue to gravity (dense air) on the leeside (down-wind)and towards the study site This wind pattern at thestudy site mimics a Katabatic wind system (Fig 5)

Katabatic wind is the technical name for a windthat carries high density air from a higher elevationdown a slope under the force of gravity Katabaticwinds originate from radiational cooling of air atop aplateau a mountain or even a hill Since the density ofair is inversely proportional to temperature the air willflow downwards warming adiabatically as it descendsThe temperature of the wind depends on thetemperature in the source region and height of descentAt Ain Sokhna it is possible that due to the presenceof this local mountain an adiabatic wind pattern existslocally and causes the air temperature to rise at thestudy site as becomes evident when one comparesthe study site air temperature observations with thosefrom Hurghada

The entire near-surface wind field and airtemperature at sea level at the study site may be heavilyinfluenced by the Katabatic winds What differentiatesthe study site from the other two stations is the factthat the mountain range is less than 500m from thestudy site and no major obstacles exist between AtakaMountain and Zakirsquos Reef (Fig 1b) The close proximityof the study site to the mountain means that elevatedtemperatures air parcels descending from Gebel Atakaare likely to cause a marked increase of air temperatureat the study site In contrast one of the main reasonsthat Hurghadarsquos temperatures are less than thoseobserved at the study site is because Ataka mountainrange is located very far (gt30 km) from the Red Seacoast For example mean observations at the studysite are very close to temperature means at Hurghada(242 vs 252oC Table 2) yet daily maximumobservations at the study site exceed Hurghada andIsmailia by almost 7 and 9 degrees respectively (Fig4) Furthermore during winter time the existence ofthe warm dry Katabatic (Foehn) wind increases airtemperature at the study site and consequently causesthe minimum daily means at the study site and Ismailiato be almost equal (Fig 4) Those warm air parcels are

1019

Int J Environ Res 9(3)1011-1022 Summer 2015

Fig 6 NOAArsquos HySplit Model results Panel A depicts forward trajectory which represents airmovement over a one week time span originating at a point approximately 10km away from ZakirsquosReef (behind the western mountain) and shows that the winds at the reefrsquos vicinity originated from

behind the western mountain all of which confirm the ascending and descending wind patternrequired for a Foehn wind Panel B depicts backward trajectory of air movement where Zakirsquos Reeflocation is the reference point This image highlights the distance traveled from the windrsquos point oforigin to the chosen endpoint (Zakirsquos Reef) and highlights the presence of a dominant westerly windin the region Both forward and backward trajectories depicted further support the existence of a

Foehn wind near the reef as a result of the consistent westerly wind and the surmounting of thepreceding mountain range

1020

Moustafa MZ et al

likely to raise water temperature particularly duringwinter season and help resident corals to survive atthis location

Warm dry Katabatic winds occur on the leeside ofa mountain range These Katabatic winds that descendon the leeside of a mountain (eg Gebel Ataka) andwarmed adiabatically are called Foehn winds (alsoChinook or Bergwind Fig 5) As a consequence of thedifferent adiabatic lapse rates of moist and dry air theair on the leeward slopes becomes warmer thanequivalent elevations on the windward slopes and canraise temperatures by as much as 30degC in just a matterof hours (Hess and McKnight 2000) For example oneof the main reasons that central Europe enjoys a warmerclimate is due to the Foehn as moisture-filled windsoff the Mediterranean Sea blow over the Alps In ourcase the end result is that conditions on the upwindside of Gebel Ataka will be relatively moist and coolwhile conditions on the leeward side of the mountaincorresponding to the study site will be relatively warmand dry As the air flow descends on the leeward sideof Gebel Ataka it becomes compressed by atmosphericpressure which raises the air-flow temperature Inaddition to compression the air flow also gains moreheat directly from the sun (particularly during summermonths) when the skies are typically clear

The increase in air temperature at sea level mayhave some positive consequences for corals at thislocation Warmer than normal air temperature due tothis local wind pattern is more likely to increase seawatertemperature at Zakirsquos Reef Comparisons between airand seawater temperature collected at Zakirsquos Reef revealthat their means are almost identical (Moustafa et al2008b 2013 2014) We speculate that the two yearlymeans of air and seawater temperatures are almostidentical as a result of the local wind pattern This warmmountain air elevates seawater temperature all of whichmay help explain the existence and survival of ZakirsquosReef at such a high latitude

HYSPLIT trajectory model was used to serve asan independent validation for the existence of thenewly identified Foehn wind Model runs ranged fromdaily to weekly traces and quantified average flow rateof the air masses and temperature rates of change asthey encountered resistances caused by naturalobstructions (eg mountain ranges) After initializingmodel runs model parameters were adjusted (egnumber of starting points temporal scale of trajectoriesand frequency of extraction) in order to observeresulting model outputs These extractions wereprojected and utilized in Google Earth to serve as avisual aid of model run outputs The resulting HYSPLITmodel data served as independent validation to supportthe proposed hypothesis regarding the existence of

Foehn winds in the Red Sea region near the study site(Fig 6)

Two HYSPLIT model run results are displayed inFig 6 The first trajectory (Fig 6a) is a forward trajectoryof air movement over a one week period originating ata point approximately 10 km West of Gabel AtakaMountain (29o 22rsquo 3047 N and 32o 15rsquo 1025 E) Thismodel run predicts that the winds in Zakirsquos Reefvicinity originate from a point behind the westwardmountain range likely confirming the risingfalling windpattern - a signature of a Foehn wind The backwardtrajectory in Fig 6b illustrates the distance traveledfrom the windrsquos point of origin to the predefined endpoint (Zakirsquos Reef) and depicts the presence of adominant westerly wind in the region These twotrajectories depicted in Fig 6 further support theexistence of a Foehn wind near Zakirsquos reef as a resultof the consistent westerly wind and the surmountingof the Ataka mountain rangeSeawater temperature is akey indicator of coral reef health Seawater temperatureinfluences ecological processes and is considered theprimary factor limiting coral reef survival anddevelopment throughout the world (Szmant ampGassman 1990 Goreau amp Hayes 2005) Seawatertemperature affects coral breeding spawning larvalsettlement and growth (Szmant amp Gassman 1990Mendes amp Woody 2002) and coral recruitment (Jokielamp Coles 1990)

It is evident that Zakirsquos coral species have arelatively higher thermal tolerance compared to similarspecies at other locations only the most hot and cold-tolerant few have been identified at this site (Moustafaet al 2008a 2013) The world-wide bleaching event in2005 resulted in lethal heat conditions to coral speciesworldwide yet coral species found at Zakirsquos reef survivetemperatures more extreme than most every yearRegardless of all exposures to lethal or near-lethal limitsof hot and cold the six coral species at Zakirsquos reefremained dominant (gt94)

Lirman et al (2011) suggested that the only coraltaxa that survive the cold event with limited mortalitywere those that are also found in other habitats thatoften experience low temperatures on a regular basissuggesting species-specific adaptive resistancemechanisms Corals on Zakirsquos reef experience theseextreme seawater temperatures annually as reportedby Moustafa et al (2014) for both summer and winterExposure to these high- and low-temperature anomalieson a yearly basis may have caused corals at Zakirsquos reefto develop an adaptive resistance mechanism It is nosurprise that only six out of the 35 identified coralspecies in the Gulf of Suez dominate Zakirsquos reef (gt94)and can be attributed to their adaptive resistancemechanisms to both low and high extreme temperatures

1021

Int J Environ Res 9(3)1011-1022 Summer 2015

These corals have been living at the study site forhundreds of years and may have the capacity toacclimatize or adapt to external stresses better thanothers For example previous comparisons amongplants (Ingram amp Bartels 1996) and animals (Huey ampKingsolver 1993 Somero 2002 Hochachka amp Somero2002) have shown that some population or species arefar more resilient compared to others with markedinfluences on growth survival and disease resistance(Nevo 2001 Hochachka amp Somero 2002 Parsons2005) Furthermore Barshis et al (2012) studied coralgenes and concluded that some of these genes maybestow resilience through faster reaction duringtransient heat stress which may prepare corals forfrequently encountered stress

Coles and Fadlallah (1991) reported that Acroporapharaonis and Porites compressa in the Arabian Gulfsuffered severe mortality when the water temperaturefell below 115oC Furthermore they reported thatPorites compressa (the dominant species at their reef)and Faviidae corals showed sign of sub-lethal effectswhen temperature was below 12oC for two consecutivedays yet recovered after six months Some of the coralgenera found in the Arabian Gulf are also found inZakirsquos reef (eg Acropora humilis and Poritescolumna) Unlike the Arabian Gulf counterpartresident Acropora in Zakirsquos reef survived well duringthe worldwide bleaching event in 2005 as well asduring cold temperatures measured in the 2006 and2007 winter seasons (Moustafa et al 2014)

CONCLUSIONSThe current study was driven mainly by scientific

curiosity regarding the discovery of a fringing coralreef ecosystem that thrived at one of the highestlatitudes of any sub-tropical reef system in the worldWe quickly learned that there are no significantenvironmental databases for this region except for thevery different Gulf of Aqaba We therefore integratedother environmental data to help us find out why thisecosystem is flourishing at such a high latitude

Our data collection and analysis lead to severalnote-worthy discoveries most importantly theexistence of a KatabaticFoehn wind system causedby the co-existence of the high mountain range andthe dominant trade winds at the study site This windsystem causes elevated temperatures during the hotsummer and winter months at the study site Thislocation is the only one where the mountains are veryclose to the coast along the entire Egyptian Red SeaAs air temperature increases water temperature at thestudy site may also increase Consequentlytemperature rise particularly during winter seasonsenhances Zakirsquos Reefrsquos chances of survival at such

high latitude Furthermore this warm local wind mayhave also helped the corals at this location evolveacclimate to the extreme conditions prevalent duringsummer months The entire Gulf of Suez fringing coralreef ecosystem may be reliant for survival particularlyduring winter months on this newly discovered localwind pattern

ACKNOWLEDGEMENTSSpecial thanks to all family members particularly

my son Zachariah Moustafa for providing fieldassistance and being the cause for expanding ourresearch In 2007 Zachariahrsquos expression caught in aphotograph from the incredible summer heat at thestudy site (registered at 135 Fahrenheit) caused us tore-evaluate our plans and start collecting meteorologicaldata Catherine Miller from Dreyfoos School of the Artsprovided lab equipment and helped find funding It iswith great appreciations to PADI and North Carolinaresearch consortium for their financial support of thisresearch Without the support of Onset and YSICorporation graciously donating and loaning fieldequipment the culmination of this research would notbeen possible particularly the dedicated staff especiallyKory Wagner from Onset and their technical supportthroughout this research project

REFERENCES

Stein A F Isakov V Godowitch J and Draxler R R(2007) A hybrid modeling approach to resolve pollutantconcentrations in an urban area Atmospheric Environment41 (40) 9410-9428

Barshis DJ Lander JT Oliver TA Seneca FO Taylor-knowles N and Palumbi SR (2012) Genomic basis forcoral resilience to climate change Proceedings of the NationalAcademy of Sciences of the United States of America 110(4) 1387-1392

Coles SL and Fadlallah YH (1991) Reef coral survivaland mortality at low temperatures in the Arabian Gulf Newspecies-specific lower temperature limits Coral Reefs 9(4)231-237

Draxler R R and Rolph G D (2013) HYSPLIT (HYbridSingle-Particle Lagrangian Integrated Trajectory) Modelaccess via NOAA ARL READY Website httpreadyarlnoaagovHYSPLITphp NOAA Air ResourcesLaboratory Silver Spring MD

Jokiel P L and Coles S L (1990) Response of Hawaiianand other Indo-Pacific reef corals to elevated temperatureCoral Reefs 8 155-162

Goreau TJF and Hayes RL (2005) Global coral reefbleaching and sea surface temperature trends from satellite-derived hotspot analysis World Resource Review 17(2) 254-293

Hanaur E 1(988) The Egyptian Red Sea A Diverrsquos Guide(San Diego CA Watersport Publishing Inc)

1018

Survival of Fringing Corals in Extreme Temperatures

rainfall and lack of a major source of fresh water resultin evaporation rates as high as 205 cmyr and highsalinity with minimal seasonal variation (Red Sea 2011)

It should be noted that of the approximately 35coral taxa known to survive in the Gulf of Suez onlysix (Acropora humilis A microclados A hemprichiiLitophyton arboretum Stylophora pistillata Poritescolumna and P plantulata (approximately 80 hardcorals and 20 soft corals) compose 94 of the reefrsquoscoral cover Coral dominance shifted over a four-yearperiod (2004-2007) mainly due to an oil spill in 2005global bleaching event (2005) and high (gt30oC) andlow (lt16oC) seawater temperature (Moustafa et al2008) Yet the six corals dominance remainedunchanged These corals regularly experience 4-65oCdaily changes in seawater temperature and seasonalvariations that exceed 29oC (Moustafa et al 2008)

There are no long-term studies in this region priorto the research reported here (Moustafa et al 2008a)This multi-year study set out to study the health ofZakirsquos Reef and its dominant coral species The factthat only six of the identified Gulf of Suez 35-40 coralspecies remained dominant for a five-year study(Moustafa et al 2008a) made us wonder what it wasabout these six species that allowed them to survive insuch a harsh environment Are these six species heattolerant or is there a linkage between fringing coralreefs located in the Gulf of Suez and the local climateInsight into local climate may provide a betterunderstanding of the direct influence weather has ifany on fringing coral reef health

What was found regarding air temperature at thestudy site and the two other locations was puzzling atfirst The fact that daily mean air temperature at thestudy site was less than the mean observed at Ismailiawas expected since the climatic trend in Egypt is thatair temperature increases going farther south Howeverit was not expected that air temperature daily means atthe study site and Hurghada would be very similar(Table 2) particularly if one considers how far Hurghadais from the study site (gt400 km south) Air temperaturedaily maximums at the study site exceeded bothHurghadarsquos and Ismailiarsquos by almost 7 and 9oCrespectively (Fig 4) While air temperature minimumdaily means are almost equal to Ismailiarsquos theminimums were less than Hurghadarsquos by 4 degreesThese observed trends were not expected consideringthe three stationsrsquo geographic locations suggestingthat other natural phenomena may be influencing airtemperature at the study site

Ain Sokhna Arabic for ldquohot springrdquo was namedafter the nearby sulfur springs These springs originateat Gebel Ataka (lt500 m West of Zakirsquos Reef) the

northern most mountains in the eastern desert Whenthe year-round trade wind approaches the study areait finds its first obstacle Gebel Ataka the mountainrange that starts at the study site and continues almostall the way south to Hurghada with an averageelevation of approximately 300m Gebel Ataka was thefirst factor that we believed may be influencing thelocal climate We hypothesized that these highplateauing mountains force the trade winds upwardand as a result it loses its moisture content anddecreases in temperature Air parcels reaching thehighest point on top of Gebel Ataka start descendingdue to gravity (dense air) on the leeside (down-wind)and towards the study site This wind pattern at thestudy site mimics a Katabatic wind system (Fig 5)

Katabatic wind is the technical name for a windthat carries high density air from a higher elevationdown a slope under the force of gravity Katabaticwinds originate from radiational cooling of air atop aplateau a mountain or even a hill Since the density ofair is inversely proportional to temperature the air willflow downwards warming adiabatically as it descendsThe temperature of the wind depends on thetemperature in the source region and height of descentAt Ain Sokhna it is possible that due to the presenceof this local mountain an adiabatic wind pattern existslocally and causes the air temperature to rise at thestudy site as becomes evident when one comparesthe study site air temperature observations with thosefrom Hurghada

The entire near-surface wind field and airtemperature at sea level at the study site may be heavilyinfluenced by the Katabatic winds What differentiatesthe study site from the other two stations is the factthat the mountain range is less than 500m from thestudy site and no major obstacles exist between AtakaMountain and Zakirsquos Reef (Fig 1b) The close proximityof the study site to the mountain means that elevatedtemperatures air parcels descending from Gebel Atakaare likely to cause a marked increase of air temperatureat the study site In contrast one of the main reasonsthat Hurghadarsquos temperatures are less than thoseobserved at the study site is because Ataka mountainrange is located very far (gt30 km) from the Red Seacoast For example mean observations at the studysite are very close to temperature means at Hurghada(242 vs 252oC Table 2) yet daily maximumobservations at the study site exceed Hurghada andIsmailia by almost 7 and 9 degrees respectively (Fig4) Furthermore during winter time the existence ofthe warm dry Katabatic (Foehn) wind increases airtemperature at the study site and consequently causesthe minimum daily means at the study site and Ismailiato be almost equal (Fig 4) Those warm air parcels are

1019

Int J Environ Res 9(3)1011-1022 Summer 2015

Fig 6 NOAArsquos HySplit Model results Panel A depicts forward trajectory which represents airmovement over a one week time span originating at a point approximately 10km away from ZakirsquosReef (behind the western mountain) and shows that the winds at the reefrsquos vicinity originated from

behind the western mountain all of which confirm the ascending and descending wind patternrequired for a Foehn wind Panel B depicts backward trajectory of air movement where Zakirsquos Reeflocation is the reference point This image highlights the distance traveled from the windrsquos point oforigin to the chosen endpoint (Zakirsquos Reef) and highlights the presence of a dominant westerly windin the region Both forward and backward trajectories depicted further support the existence of a

Foehn wind near the reef as a result of the consistent westerly wind and the surmounting of thepreceding mountain range

1020

Moustafa MZ et al

likely to raise water temperature particularly duringwinter season and help resident corals to survive atthis location

Warm dry Katabatic winds occur on the leeside ofa mountain range These Katabatic winds that descendon the leeside of a mountain (eg Gebel Ataka) andwarmed adiabatically are called Foehn winds (alsoChinook or Bergwind Fig 5) As a consequence of thedifferent adiabatic lapse rates of moist and dry air theair on the leeward slopes becomes warmer thanequivalent elevations on the windward slopes and canraise temperatures by as much as 30degC in just a matterof hours (Hess and McKnight 2000) For example oneof the main reasons that central Europe enjoys a warmerclimate is due to the Foehn as moisture-filled windsoff the Mediterranean Sea blow over the Alps In ourcase the end result is that conditions on the upwindside of Gebel Ataka will be relatively moist and coolwhile conditions on the leeward side of the mountaincorresponding to the study site will be relatively warmand dry As the air flow descends on the leeward sideof Gebel Ataka it becomes compressed by atmosphericpressure which raises the air-flow temperature Inaddition to compression the air flow also gains moreheat directly from the sun (particularly during summermonths) when the skies are typically clear

The increase in air temperature at sea level mayhave some positive consequences for corals at thislocation Warmer than normal air temperature due tothis local wind pattern is more likely to increase seawatertemperature at Zakirsquos Reef Comparisons between airand seawater temperature collected at Zakirsquos Reef revealthat their means are almost identical (Moustafa et al2008b 2013 2014) We speculate that the two yearlymeans of air and seawater temperatures are almostidentical as a result of the local wind pattern This warmmountain air elevates seawater temperature all of whichmay help explain the existence and survival of ZakirsquosReef at such a high latitude

HYSPLIT trajectory model was used to serve asan independent validation for the existence of thenewly identified Foehn wind Model runs ranged fromdaily to weekly traces and quantified average flow rateof the air masses and temperature rates of change asthey encountered resistances caused by naturalobstructions (eg mountain ranges) After initializingmodel runs model parameters were adjusted (egnumber of starting points temporal scale of trajectoriesand frequency of extraction) in order to observeresulting model outputs These extractions wereprojected and utilized in Google Earth to serve as avisual aid of model run outputs The resulting HYSPLITmodel data served as independent validation to supportthe proposed hypothesis regarding the existence of

Foehn winds in the Red Sea region near the study site(Fig 6)

Two HYSPLIT model run results are displayed inFig 6 The first trajectory (Fig 6a) is a forward trajectoryof air movement over a one week period originating ata point approximately 10 km West of Gabel AtakaMountain (29o 22rsquo 3047 N and 32o 15rsquo 1025 E) Thismodel run predicts that the winds in Zakirsquos Reefvicinity originate from a point behind the westwardmountain range likely confirming the risingfalling windpattern - a signature of a Foehn wind The backwardtrajectory in Fig 6b illustrates the distance traveledfrom the windrsquos point of origin to the predefined endpoint (Zakirsquos Reef) and depicts the presence of adominant westerly wind in the region These twotrajectories depicted in Fig 6 further support theexistence of a Foehn wind near Zakirsquos reef as a resultof the consistent westerly wind and the surmountingof the Ataka mountain rangeSeawater temperature is akey indicator of coral reef health Seawater temperatureinfluences ecological processes and is considered theprimary factor limiting coral reef survival anddevelopment throughout the world (Szmant ampGassman 1990 Goreau amp Hayes 2005) Seawatertemperature affects coral breeding spawning larvalsettlement and growth (Szmant amp Gassman 1990Mendes amp Woody 2002) and coral recruitment (Jokielamp Coles 1990)

It is evident that Zakirsquos coral species have arelatively higher thermal tolerance compared to similarspecies at other locations only the most hot and cold-tolerant few have been identified at this site (Moustafaet al 2008a 2013) The world-wide bleaching event in2005 resulted in lethal heat conditions to coral speciesworldwide yet coral species found at Zakirsquos reef survivetemperatures more extreme than most every yearRegardless of all exposures to lethal or near-lethal limitsof hot and cold the six coral species at Zakirsquos reefremained dominant (gt94)

Lirman et al (2011) suggested that the only coraltaxa that survive the cold event with limited mortalitywere those that are also found in other habitats thatoften experience low temperatures on a regular basissuggesting species-specific adaptive resistancemechanisms Corals on Zakirsquos reef experience theseextreme seawater temperatures annually as reportedby Moustafa et al (2014) for both summer and winterExposure to these high- and low-temperature anomalieson a yearly basis may have caused corals at Zakirsquos reefto develop an adaptive resistance mechanism It is nosurprise that only six out of the 35 identified coralspecies in the Gulf of Suez dominate Zakirsquos reef (gt94)and can be attributed to their adaptive resistancemechanisms to both low and high extreme temperatures

1021

Int J Environ Res 9(3)1011-1022 Summer 2015

These corals have been living at the study site forhundreds of years and may have the capacity toacclimatize or adapt to external stresses better thanothers For example previous comparisons amongplants (Ingram amp Bartels 1996) and animals (Huey ampKingsolver 1993 Somero 2002 Hochachka amp Somero2002) have shown that some population or species arefar more resilient compared to others with markedinfluences on growth survival and disease resistance(Nevo 2001 Hochachka amp Somero 2002 Parsons2005) Furthermore Barshis et al (2012) studied coralgenes and concluded that some of these genes maybestow resilience through faster reaction duringtransient heat stress which may prepare corals forfrequently encountered stress

Coles and Fadlallah (1991) reported that Acroporapharaonis and Porites compressa in the Arabian Gulfsuffered severe mortality when the water temperaturefell below 115oC Furthermore they reported thatPorites compressa (the dominant species at their reef)and Faviidae corals showed sign of sub-lethal effectswhen temperature was below 12oC for two consecutivedays yet recovered after six months Some of the coralgenera found in the Arabian Gulf are also found inZakirsquos reef (eg Acropora humilis and Poritescolumna) Unlike the Arabian Gulf counterpartresident Acropora in Zakirsquos reef survived well duringthe worldwide bleaching event in 2005 as well asduring cold temperatures measured in the 2006 and2007 winter seasons (Moustafa et al 2014)

CONCLUSIONSThe current study was driven mainly by scientific

curiosity regarding the discovery of a fringing coralreef ecosystem that thrived at one of the highestlatitudes of any sub-tropical reef system in the worldWe quickly learned that there are no significantenvironmental databases for this region except for thevery different Gulf of Aqaba We therefore integratedother environmental data to help us find out why thisecosystem is flourishing at such a high latitude

Our data collection and analysis lead to severalnote-worthy discoveries most importantly theexistence of a KatabaticFoehn wind system causedby the co-existence of the high mountain range andthe dominant trade winds at the study site This windsystem causes elevated temperatures during the hotsummer and winter months at the study site Thislocation is the only one where the mountains are veryclose to the coast along the entire Egyptian Red SeaAs air temperature increases water temperature at thestudy site may also increase Consequentlytemperature rise particularly during winter seasonsenhances Zakirsquos Reefrsquos chances of survival at such

high latitude Furthermore this warm local wind mayhave also helped the corals at this location evolveacclimate to the extreme conditions prevalent duringsummer months The entire Gulf of Suez fringing coralreef ecosystem may be reliant for survival particularlyduring winter months on this newly discovered localwind pattern

ACKNOWLEDGEMENTSSpecial thanks to all family members particularly

my son Zachariah Moustafa for providing fieldassistance and being the cause for expanding ourresearch In 2007 Zachariahrsquos expression caught in aphotograph from the incredible summer heat at thestudy site (registered at 135 Fahrenheit) caused us tore-evaluate our plans and start collecting meteorologicaldata Catherine Miller from Dreyfoos School of the Artsprovided lab equipment and helped find funding It iswith great appreciations to PADI and North Carolinaresearch consortium for their financial support of thisresearch Without the support of Onset and YSICorporation graciously donating and loaning fieldequipment the culmination of this research would notbeen possible particularly the dedicated staff especiallyKory Wagner from Onset and their technical supportthroughout this research project

REFERENCES

Stein A F Isakov V Godowitch J and Draxler R R(2007) A hybrid modeling approach to resolve pollutantconcentrations in an urban area Atmospheric Environment41 (40) 9410-9428

Barshis DJ Lander JT Oliver TA Seneca FO Taylor-knowles N and Palumbi SR (2012) Genomic basis forcoral resilience to climate change Proceedings of the NationalAcademy of Sciences of the United States of America 110(4) 1387-1392

Coles SL and Fadlallah YH (1991) Reef coral survivaland mortality at low temperatures in the Arabian Gulf Newspecies-specific lower temperature limits Coral Reefs 9(4)231-237

Draxler R R and Rolph G D (2013) HYSPLIT (HYbridSingle-Particle Lagrangian Integrated Trajectory) Modelaccess via NOAA ARL READY Website httpreadyarlnoaagovHYSPLITphp NOAA Air ResourcesLaboratory Silver Spring MD

Jokiel P L and Coles S L (1990) Response of Hawaiianand other Indo-Pacific reef corals to elevated temperatureCoral Reefs 8 155-162

Goreau TJF and Hayes RL (2005) Global coral reefbleaching and sea surface temperature trends from satellite-derived hotspot analysis World Resource Review 17(2) 254-293

Hanaur E 1(988) The Egyptian Red Sea A Diverrsquos Guide(San Diego CA Watersport Publishing Inc)

1019

Int J Environ Res 9(3)1011-1022 Summer 2015

Fig 6 NOAArsquos HySplit Model results Panel A depicts forward trajectory which represents airmovement over a one week time span originating at a point approximately 10km away from ZakirsquosReef (behind the western mountain) and shows that the winds at the reefrsquos vicinity originated from

behind the western mountain all of which confirm the ascending and descending wind patternrequired for a Foehn wind Panel B depicts backward trajectory of air movement where Zakirsquos Reeflocation is the reference point This image highlights the distance traveled from the windrsquos point oforigin to the chosen endpoint (Zakirsquos Reef) and highlights the presence of a dominant westerly windin the region Both forward and backward trajectories depicted further support the existence of a

Foehn wind near the reef as a result of the consistent westerly wind and the surmounting of thepreceding mountain range

1020

Moustafa MZ et al

likely to raise water temperature particularly duringwinter season and help resident corals to survive atthis location

Warm dry Katabatic winds occur on the leeside ofa mountain range These Katabatic winds that descendon the leeside of a mountain (eg Gebel Ataka) andwarmed adiabatically are called Foehn winds (alsoChinook or Bergwind Fig 5) As a consequence of thedifferent adiabatic lapse rates of moist and dry air theair on the leeward slopes becomes warmer thanequivalent elevations on the windward slopes and canraise temperatures by as much as 30degC in just a matterof hours (Hess and McKnight 2000) For example oneof the main reasons that central Europe enjoys a warmerclimate is due to the Foehn as moisture-filled windsoff the Mediterranean Sea blow over the Alps In ourcase the end result is that conditions on the upwindside of Gebel Ataka will be relatively moist and coolwhile conditions on the leeward side of the mountaincorresponding to the study site will be relatively warmand dry As the air flow descends on the leeward sideof Gebel Ataka it becomes compressed by atmosphericpressure which raises the air-flow temperature Inaddition to compression the air flow also gains moreheat directly from the sun (particularly during summermonths) when the skies are typically clear

The increase in air temperature at sea level mayhave some positive consequences for corals at thislocation Warmer than normal air temperature due tothis local wind pattern is more likely to increase seawatertemperature at Zakirsquos Reef Comparisons between airand seawater temperature collected at Zakirsquos Reef revealthat their means are almost identical (Moustafa et al2008b 2013 2014) We speculate that the two yearlymeans of air and seawater temperatures are almostidentical as a result of the local wind pattern This warmmountain air elevates seawater temperature all of whichmay help explain the existence and survival of ZakirsquosReef at such a high latitude

HYSPLIT trajectory model was used to serve asan independent validation for the existence of thenewly identified Foehn wind Model runs ranged fromdaily to weekly traces and quantified average flow rateof the air masses and temperature rates of change asthey encountered resistances caused by naturalobstructions (eg mountain ranges) After initializingmodel runs model parameters were adjusted (egnumber of starting points temporal scale of trajectoriesand frequency of extraction) in order to observeresulting model outputs These extractions wereprojected and utilized in Google Earth to serve as avisual aid of model run outputs The resulting HYSPLITmodel data served as independent validation to supportthe proposed hypothesis regarding the existence of

Foehn winds in the Red Sea region near the study site(Fig 6)

Two HYSPLIT model run results are displayed inFig 6 The first trajectory (Fig 6a) is a forward trajectoryof air movement over a one week period originating ata point approximately 10 km West of Gabel AtakaMountain (29o 22rsquo 3047 N and 32o 15rsquo 1025 E) Thismodel run predicts that the winds in Zakirsquos Reefvicinity originate from a point behind the westwardmountain range likely confirming the risingfalling windpattern - a signature of a Foehn wind The backwardtrajectory in Fig 6b illustrates the distance traveledfrom the windrsquos point of origin to the predefined endpoint (Zakirsquos Reef) and depicts the presence of adominant westerly wind in the region These twotrajectories depicted in Fig 6 further support theexistence of a Foehn wind near Zakirsquos reef as a resultof the consistent westerly wind and the surmountingof the Ataka mountain rangeSeawater temperature is akey indicator of coral reef health Seawater temperatureinfluences ecological processes and is considered theprimary factor limiting coral reef survival anddevelopment throughout the world (Szmant ampGassman 1990 Goreau amp Hayes 2005) Seawatertemperature affects coral breeding spawning larvalsettlement and growth (Szmant amp Gassman 1990Mendes amp Woody 2002) and coral recruitment (Jokielamp Coles 1990)

It is evident that Zakirsquos coral species have arelatively higher thermal tolerance compared to similarspecies at other locations only the most hot and cold-tolerant few have been identified at this site (Moustafaet al 2008a 2013) The world-wide bleaching event in2005 resulted in lethal heat conditions to coral speciesworldwide yet coral species found at Zakirsquos reef survivetemperatures more extreme than most every yearRegardless of all exposures to lethal or near-lethal limitsof hot and cold the six coral species at Zakirsquos reefremained dominant (gt94)

Lirman et al (2011) suggested that the only coraltaxa that survive the cold event with limited mortalitywere those that are also found in other habitats thatoften experience low temperatures on a regular basissuggesting species-specific adaptive resistancemechanisms Corals on Zakirsquos reef experience theseextreme seawater temperatures annually as reportedby Moustafa et al (2014) for both summer and winterExposure to these high- and low-temperature anomalieson a yearly basis may have caused corals at Zakirsquos reefto develop an adaptive resistance mechanism It is nosurprise that only six out of the 35 identified coralspecies in the Gulf of Suez dominate Zakirsquos reef (gt94)and can be attributed to their adaptive resistancemechanisms to both low and high extreme temperatures

1021

Int J Environ Res 9(3)1011-1022 Summer 2015

These corals have been living at the study site forhundreds of years and may have the capacity toacclimatize or adapt to external stresses better thanothers For example previous comparisons amongplants (Ingram amp Bartels 1996) and animals (Huey ampKingsolver 1993 Somero 2002 Hochachka amp Somero2002) have shown that some population or species arefar more resilient compared to others with markedinfluences on growth survival and disease resistance(Nevo 2001 Hochachka amp Somero 2002 Parsons2005) Furthermore Barshis et al (2012) studied coralgenes and concluded that some of these genes maybestow resilience through faster reaction duringtransient heat stress which may prepare corals forfrequently encountered stress

Coles and Fadlallah (1991) reported that Acroporapharaonis and Porites compressa in the Arabian Gulfsuffered severe mortality when the water temperaturefell below 115oC Furthermore they reported thatPorites compressa (the dominant species at their reef)and Faviidae corals showed sign of sub-lethal effectswhen temperature was below 12oC for two consecutivedays yet recovered after six months Some of the coralgenera found in the Arabian Gulf are also found inZakirsquos reef (eg Acropora humilis and Poritescolumna) Unlike the Arabian Gulf counterpartresident Acropora in Zakirsquos reef survived well duringthe worldwide bleaching event in 2005 as well asduring cold temperatures measured in the 2006 and2007 winter seasons (Moustafa et al 2014)

CONCLUSIONSThe current study was driven mainly by scientific

curiosity regarding the discovery of a fringing coralreef ecosystem that thrived at one of the highestlatitudes of any sub-tropical reef system in the worldWe quickly learned that there are no significantenvironmental databases for this region except for thevery different Gulf of Aqaba We therefore integratedother environmental data to help us find out why thisecosystem is flourishing at such a high latitude

Our data collection and analysis lead to severalnote-worthy discoveries most importantly theexistence of a KatabaticFoehn wind system causedby the co-existence of the high mountain range andthe dominant trade winds at the study site This windsystem causes elevated temperatures during the hotsummer and winter months at the study site Thislocation is the only one where the mountains are veryclose to the coast along the entire Egyptian Red SeaAs air temperature increases water temperature at thestudy site may also increase Consequentlytemperature rise particularly during winter seasonsenhances Zakirsquos Reefrsquos chances of survival at such

high latitude Furthermore this warm local wind mayhave also helped the corals at this location evolveacclimate to the extreme conditions prevalent duringsummer months The entire Gulf of Suez fringing coralreef ecosystem may be reliant for survival particularlyduring winter months on this newly discovered localwind pattern

ACKNOWLEDGEMENTSSpecial thanks to all family members particularly

my son Zachariah Moustafa for providing fieldassistance and being the cause for expanding ourresearch In 2007 Zachariahrsquos expression caught in aphotograph from the incredible summer heat at thestudy site (registered at 135 Fahrenheit) caused us tore-evaluate our plans and start collecting meteorologicaldata Catherine Miller from Dreyfoos School of the Artsprovided lab equipment and helped find funding It iswith great appreciations to PADI and North Carolinaresearch consortium for their financial support of thisresearch Without the support of Onset and YSICorporation graciously donating and loaning fieldequipment the culmination of this research would notbeen possible particularly the dedicated staff especiallyKory Wagner from Onset and their technical supportthroughout this research project

REFERENCES

Stein A F Isakov V Godowitch J and Draxler R R(2007) A hybrid modeling approach to resolve pollutantconcentrations in an urban area Atmospheric Environment41 (40) 9410-9428

Barshis DJ Lander JT Oliver TA Seneca FO Taylor-knowles N and Palumbi SR (2012) Genomic basis forcoral resilience to climate change Proceedings of the NationalAcademy of Sciences of the United States of America 110(4) 1387-1392

Coles SL and Fadlallah YH (1991) Reef coral survivaland mortality at low temperatures in the Arabian Gulf Newspecies-specific lower temperature limits Coral Reefs 9(4)231-237

Draxler R R and Rolph G D (2013) HYSPLIT (HYbridSingle-Particle Lagrangian Integrated Trajectory) Modelaccess via NOAA ARL READY Website httpreadyarlnoaagovHYSPLITphp NOAA Air ResourcesLaboratory Silver Spring MD

Jokiel P L and Coles S L (1990) Response of Hawaiianand other Indo-Pacific reef corals to elevated temperatureCoral Reefs 8 155-162

Goreau TJF and Hayes RL (2005) Global coral reefbleaching and sea surface temperature trends from satellite-derived hotspot analysis World Resource Review 17(2) 254-293

Hanaur E 1(988) The Egyptian Red Sea A Diverrsquos Guide(San Diego CA Watersport Publishing Inc)

1020

Moustafa MZ et al

likely to raise water temperature particularly duringwinter season and help resident corals to survive atthis location

Warm dry Katabatic winds occur on the leeside ofa mountain range These Katabatic winds that descendon the leeside of a mountain (eg Gebel Ataka) andwarmed adiabatically are called Foehn winds (alsoChinook or Bergwind Fig 5) As a consequence of thedifferent adiabatic lapse rates of moist and dry air theair on the leeward slopes becomes warmer thanequivalent elevations on the windward slopes and canraise temperatures by as much as 30degC in just a matterof hours (Hess and McKnight 2000) For example oneof the main reasons that central Europe enjoys a warmerclimate is due to the Foehn as moisture-filled windsoff the Mediterranean Sea blow over the Alps In ourcase the end result is that conditions on the upwindside of Gebel Ataka will be relatively moist and coolwhile conditions on the leeward side of the mountaincorresponding to the study site will be relatively warmand dry As the air flow descends on the leeward sideof Gebel Ataka it becomes compressed by atmosphericpressure which raises the air-flow temperature Inaddition to compression the air flow also gains moreheat directly from the sun (particularly during summermonths) when the skies are typically clear

The increase in air temperature at sea level mayhave some positive consequences for corals at thislocation Warmer than normal air temperature due tothis local wind pattern is more likely to increase seawatertemperature at Zakirsquos Reef Comparisons between airand seawater temperature collected at Zakirsquos Reef revealthat their means are almost identical (Moustafa et al2008b 2013 2014) We speculate that the two yearlymeans of air and seawater temperatures are almostidentical as a result of the local wind pattern This warmmountain air elevates seawater temperature all of whichmay help explain the existence and survival of ZakirsquosReef at such a high latitude

HYSPLIT trajectory model was used to serve asan independent validation for the existence of thenewly identified Foehn wind Model runs ranged fromdaily to weekly traces and quantified average flow rateof the air masses and temperature rates of change asthey encountered resistances caused by naturalobstructions (eg mountain ranges) After initializingmodel runs model parameters were adjusted (egnumber of starting points temporal scale of trajectoriesand frequency of extraction) in order to observeresulting model outputs These extractions wereprojected and utilized in Google Earth to serve as avisual aid of model run outputs The resulting HYSPLITmodel data served as independent validation to supportthe proposed hypothesis regarding the existence of

Foehn winds in the Red Sea region near the study site(Fig 6)

Two HYSPLIT model run results are displayed inFig 6 The first trajectory (Fig 6a) is a forward trajectoryof air movement over a one week period originating ata point approximately 10 km West of Gabel AtakaMountain (29o 22rsquo 3047 N and 32o 15rsquo 1025 E) Thismodel run predicts that the winds in Zakirsquos Reefvicinity originate from a point behind the westwardmountain range likely confirming the risingfalling windpattern - a signature of a Foehn wind The backwardtrajectory in Fig 6b illustrates the distance traveledfrom the windrsquos point of origin to the predefined endpoint (Zakirsquos Reef) and depicts the presence of adominant westerly wind in the region These twotrajectories depicted in Fig 6 further support theexistence of a Foehn wind near Zakirsquos reef as a resultof the consistent westerly wind and the surmountingof the Ataka mountain rangeSeawater temperature is akey indicator of coral reef health Seawater temperatureinfluences ecological processes and is considered theprimary factor limiting coral reef survival anddevelopment throughout the world (Szmant ampGassman 1990 Goreau amp Hayes 2005) Seawatertemperature affects coral breeding spawning larvalsettlement and growth (Szmant amp Gassman 1990Mendes amp Woody 2002) and coral recruitment (Jokielamp Coles 1990)

It is evident that Zakirsquos coral species have arelatively higher thermal tolerance compared to similarspecies at other locations only the most hot and cold-tolerant few have been identified at this site (Moustafaet al 2008a 2013) The world-wide bleaching event in2005 resulted in lethal heat conditions to coral speciesworldwide yet coral species found at Zakirsquos reef survivetemperatures more extreme than most every yearRegardless of all exposures to lethal or near-lethal limitsof hot and cold the six coral species at Zakirsquos reefremained dominant (gt94)

Lirman et al (2011) suggested that the only coraltaxa that survive the cold event with limited mortalitywere those that are also found in other habitats thatoften experience low temperatures on a regular basissuggesting species-specific adaptive resistancemechanisms Corals on Zakirsquos reef experience theseextreme seawater temperatures annually as reportedby Moustafa et al (2014) for both summer and winterExposure to these high- and low-temperature anomalieson a yearly basis may have caused corals at Zakirsquos reefto develop an adaptive resistance mechanism It is nosurprise that only six out of the 35 identified coralspecies in the Gulf of Suez dominate Zakirsquos reef (gt94)and can be attributed to their adaptive resistancemechanisms to both low and high extreme temperatures

1021

Int J Environ Res 9(3)1011-1022 Summer 2015

These corals have been living at the study site forhundreds of years and may have the capacity toacclimatize or adapt to external stresses better thanothers For example previous comparisons amongplants (Ingram amp Bartels 1996) and animals (Huey ampKingsolver 1993 Somero 2002 Hochachka amp Somero2002) have shown that some population or species arefar more resilient compared to others with markedinfluences on growth survival and disease resistance(Nevo 2001 Hochachka amp Somero 2002 Parsons2005) Furthermore Barshis et al (2012) studied coralgenes and concluded that some of these genes maybestow resilience through faster reaction duringtransient heat stress which may prepare corals forfrequently encountered stress

Coles and Fadlallah (1991) reported that Acroporapharaonis and Porites compressa in the Arabian Gulfsuffered severe mortality when the water temperaturefell below 115oC Furthermore they reported thatPorites compressa (the dominant species at their reef)and Faviidae corals showed sign of sub-lethal effectswhen temperature was below 12oC for two consecutivedays yet recovered after six months Some of the coralgenera found in the Arabian Gulf are also found inZakirsquos reef (eg Acropora humilis and Poritescolumna) Unlike the Arabian Gulf counterpartresident Acropora in Zakirsquos reef survived well duringthe worldwide bleaching event in 2005 as well asduring cold temperatures measured in the 2006 and2007 winter seasons (Moustafa et al 2014)

CONCLUSIONSThe current study was driven mainly by scientific

curiosity regarding the discovery of a fringing coralreef ecosystem that thrived at one of the highestlatitudes of any sub-tropical reef system in the worldWe quickly learned that there are no significantenvironmental databases for this region except for thevery different Gulf of Aqaba We therefore integratedother environmental data to help us find out why thisecosystem is flourishing at such a high latitude

Our data collection and analysis lead to severalnote-worthy discoveries most importantly theexistence of a KatabaticFoehn wind system causedby the co-existence of the high mountain range andthe dominant trade winds at the study site This windsystem causes elevated temperatures during the hotsummer and winter months at the study site Thislocation is the only one where the mountains are veryclose to the coast along the entire Egyptian Red SeaAs air temperature increases water temperature at thestudy site may also increase Consequentlytemperature rise particularly during winter seasonsenhances Zakirsquos Reefrsquos chances of survival at such

high latitude Furthermore this warm local wind mayhave also helped the corals at this location evolveacclimate to the extreme conditions prevalent duringsummer months The entire Gulf of Suez fringing coralreef ecosystem may be reliant for survival particularlyduring winter months on this newly discovered localwind pattern

ACKNOWLEDGEMENTSSpecial thanks to all family members particularly

my son Zachariah Moustafa for providing fieldassistance and being the cause for expanding ourresearch In 2007 Zachariahrsquos expression caught in aphotograph from the incredible summer heat at thestudy site (registered at 135 Fahrenheit) caused us tore-evaluate our plans and start collecting meteorologicaldata Catherine Miller from Dreyfoos School of the Artsprovided lab equipment and helped find funding It iswith great appreciations to PADI and North Carolinaresearch consortium for their financial support of thisresearch Without the support of Onset and YSICorporation graciously donating and loaning fieldequipment the culmination of this research would notbeen possible particularly the dedicated staff especiallyKory Wagner from Onset and their technical supportthroughout this research project

REFERENCES

Stein A F Isakov V Godowitch J and Draxler R R(2007) A hybrid modeling approach to resolve pollutantconcentrations in an urban area Atmospheric Environment41 (40) 9410-9428

Barshis DJ Lander JT Oliver TA Seneca FO Taylor-knowles N and Palumbi SR (2012) Genomic basis forcoral resilience to climate change Proceedings of the NationalAcademy of Sciences of the United States of America 110(4) 1387-1392

Coles SL and Fadlallah YH (1991) Reef coral survivaland mortality at low temperatures in the Arabian Gulf Newspecies-specific lower temperature limits Coral Reefs 9(4)231-237

Draxler R R and Rolph G D (2013) HYSPLIT (HYbridSingle-Particle Lagrangian Integrated Trajectory) Modelaccess via NOAA ARL READY Website httpreadyarlnoaagovHYSPLITphp NOAA Air ResourcesLaboratory Silver Spring MD

Jokiel P L and Coles S L (1990) Response of Hawaiianand other Indo-Pacific reef corals to elevated temperatureCoral Reefs 8 155-162

Goreau TJF and Hayes RL (2005) Global coral reefbleaching and sea surface temperature trends from satellite-derived hotspot analysis World Resource Review 17(2) 254-293

Hanaur E 1(988) The Egyptian Red Sea A Diverrsquos Guide(San Diego CA Watersport Publishing Inc)

1021

Int J Environ Res 9(3)1011-1022 Summer 2015

These corals have been living at the study site forhundreds of years and may have the capacity toacclimatize or adapt to external stresses better thanothers For example previous comparisons amongplants (Ingram amp Bartels 1996) and animals (Huey ampKingsolver 1993 Somero 2002 Hochachka amp Somero2002) have shown that some population or species arefar more resilient compared to others with markedinfluences on growth survival and disease resistance(Nevo 2001 Hochachka amp Somero 2002 Parsons2005) Furthermore Barshis et al (2012) studied coralgenes and concluded that some of these genes maybestow resilience through faster reaction duringtransient heat stress which may prepare corals forfrequently encountered stress

Coles and Fadlallah (1991) reported that Acroporapharaonis and Porites compressa in the Arabian Gulfsuffered severe mortality when the water temperaturefell below 115oC Furthermore they reported thatPorites compressa (the dominant species at their reef)and Faviidae corals showed sign of sub-lethal effectswhen temperature was below 12oC for two consecutivedays yet recovered after six months Some of the coralgenera found in the Arabian Gulf are also found inZakirsquos reef (eg Acropora humilis and Poritescolumna) Unlike the Arabian Gulf counterpartresident Acropora in Zakirsquos reef survived well duringthe worldwide bleaching event in 2005 as well asduring cold temperatures measured in the 2006 and2007 winter seasons (Moustafa et al 2014)

CONCLUSIONSThe current study was driven mainly by scientific

curiosity regarding the discovery of a fringing coralreef ecosystem that thrived at one of the highestlatitudes of any sub-tropical reef system in the worldWe quickly learned that there are no significantenvironmental databases for this region except for thevery different Gulf of Aqaba We therefore integratedother environmental data to help us find out why thisecosystem is flourishing at such a high latitude

Our data collection and analysis lead to severalnote-worthy discoveries most importantly theexistence of a KatabaticFoehn wind system causedby the co-existence of the high mountain range andthe dominant trade winds at the study site This windsystem causes elevated temperatures during the hotsummer and winter months at the study site Thislocation is the only one where the mountains are veryclose to the coast along the entire Egyptian Red SeaAs air temperature increases water temperature at thestudy site may also increase Consequentlytemperature rise particularly during winter seasonsenhances Zakirsquos Reefrsquos chances of survival at such

high latitude Furthermore this warm local wind mayhave also helped the corals at this location evolveacclimate to the extreme conditions prevalent duringsummer months The entire Gulf of Suez fringing coralreef ecosystem may be reliant for survival particularlyduring winter months on this newly discovered localwind pattern

ACKNOWLEDGEMENTSSpecial thanks to all family members particularly

my son Zachariah Moustafa for providing fieldassistance and being the cause for expanding ourresearch In 2007 Zachariahrsquos expression caught in aphotograph from the incredible summer heat at thestudy site (registered at 135 Fahrenheit) caused us tore-evaluate our plans and start collecting meteorologicaldata Catherine Miller from Dreyfoos School of the Artsprovided lab equipment and helped find funding It iswith great appreciations to PADI and North Carolinaresearch consortium for their financial support of thisresearch Without the support of Onset and YSICorporation graciously donating and loaning fieldequipment the culmination of this research would notbeen possible particularly the dedicated staff especiallyKory Wagner from Onset and their technical supportthroughout this research project

REFERENCES

Stein A F Isakov V Godowitch J and Draxler R R(2007) A hybrid modeling approach to resolve pollutantconcentrations in an urban area Atmospheric Environment41 (40) 9410-9428

Barshis DJ Lander JT Oliver TA Seneca FO Taylor-knowles N and Palumbi SR (2012) Genomic basis forcoral resilience to climate change Proceedings of the NationalAcademy of Sciences of the United States of America 110(4) 1387-1392

Coles SL and Fadlallah YH (1991) Reef coral survivaland mortality at low temperatures in the Arabian Gulf Newspecies-specific lower temperature limits Coral Reefs 9(4)231-237

Draxler R R and Rolph G D (2013) HYSPLIT (HYbridSingle-Particle Lagrangian Integrated Trajectory) Modelaccess via NOAA ARL READY Website httpreadyarlnoaagovHYSPLITphp NOAA Air ResourcesLaboratory Silver Spring MD

Jokiel P L and Coles S L (1990) Response of Hawaiianand other Indo-Pacific reef corals to elevated temperatureCoral Reefs 8 155-162

Goreau TJF and Hayes RL (2005) Global coral reefbleaching and sea surface temperature trends from satellite-derived hotspot analysis World Resource Review 17(2) 254-293

Hanaur E 1(988) The Egyptian Red Sea A Diverrsquos Guide(San Diego CA Watersport Publishing Inc)

1022

Survival of Fringing Corals in Extreme Temperatures

Hess D and McKnight TL (2000) FoehnChinook WindsIn Physical Geography A Landscape Appreciation UpperSaddle River Prentice Hall NJ

Hochachka P W and Somero G N (2002) BiochemicalAdaptation Mechanism and Process in PhysiologicalEvolution (New York Oxford Univ Press)

Huey R B and Kingsolver JG (1993) Evolution ofresistance to high-temperatures in ectotherms Am Nat 142(Suppl) S21ndashS46

Ingram J and Bartels D (1996) The molecular basis ofdehydration tolerance in plants Annu Rev Plant PhysiolPlant Mol Biol 47 377ndash403

Lirman D Schopmeyer S Manzello D Gramer LJPrecht WF et al (2011) Severe Cold-Water Event CausedUnprecedented Mortality to Corals of the Florida Reef Tractand Reversed Previous Survivorship Patterns PLoS ONE6(8) e23047

Mendes J M and Woodley JD (2002) Effect of the 1995-1996 bleaching event on polyp tissue depth growthreproduction and skeletal band formation in Montastraeaannularis Mar Ecol Prog Ser 235 93ndash102

Morcos S A (1970) Physical and chemical oceanographyof the Red Sea Oceanogr Mar Bio Ann Rev 8 73-202

Mustafa F (2000) Status of Coral Reefs in the Middle EastAIMS httpwwwaimsgovau Accessed 62007

Moustafa Z D Hallock P Moustafa M S and MoustafaM Z (2008a) Survivalship of a Red Sea Fringing CoralReef under Extreme Environmental Conditions Proc of the11th International Coral Reef Symposium Ft LauderdaleFlorida

Moustafa Z D Moustafa M S and Moustafa M Z(2008b) What is Normal Extreme Temperature Variabilityon a High Latitude Fringing Red Sea Coral Reefrdquo ASLOAGU Ocean Science Meeting Orlando FL

Moustafa M Z Moustafa M S Moustafa Z D andMoustafa SE (2014) Survival of high latitude fringing coralsin extreme temperatures Red Sea Oceanography Journal ofsea research 88 144-151

Moustafa M Z Moustafa Z D and Moustafa M S(2013) Resilience of high latitude Res Sea Corals to extremetemperature OJE 3(3) 242-253 doi 104236oje201333028

Nevo E (2001) Evolution of genome-phenome diversityunder environmental stress Proc Natl Acad Sci 98(11)6233ndash6240

Okamoto K and Derber JC (2006) Assimilation of SSMI Radiances in the NCEP Global Data Assimilation SystemMonthly Weather Review 134 2612-2631 httpreadyarlnoaagovgdas1php

Onset Computer Corporation (2014) Onset HOBO DataLoggers Product Catalog Data Loggers External Sensorsand Software httpwwwonsetcompcomfilescatalog-2014-lrpdf

Parsons PA (2005) Environments and evolutionInteractions between stress resource inadequacy andenergetic efficiency Biol Rev Camb Philos Soc 80 (4) 589ndash610

Red Sea (2011) In Encyclopedia Britannica Retrieved June11 2011 from Encyclopedia Britannica Online httpwwwbritannicacomEBcheckedtopic494479Red-Sea

Shahin M (1989) Review and assessment of water resourcesin the Arab region Water Int 14 206ndash219

Somero G N (2002) Thermal physiology and verticalzonation of intertidal animals Optima limits and costs ofliving Integr Comp Biol 42(4) 780ndash789

Szmant A M and Gassman N J (1990) The effects ofprolonged lsquolsquobleaching rsquorsquo on the tissue biomass andreproduction of the reef coral Montastrea annularis CoralReefs 8 217ndash224

UNESCO (2005) Regional Workshop on Impact ofEcotourism on Biosphere reserves Sharm El Sheikh Egypt13-15 November 2005 httpwwwunescoorgnewf i leadm in M U LT IM ED IA FIE LD C a ir o im ages FinalReport_Ecotourismpdf

Wunderground (2010) Weather data web site (httpwwwwundergroundcomglobalEGhtml)