Designing a Diving Protocol for Thermocline Identification Using

applied sciences

Article Designing a Diving Protocol for Thermocline Identiﬁcation Using Dive Computers in Marine Citizen Science

Salih Murat Egi 1,2 , Pierre-Yves Cousteau 3, Massimo Pieri 1, Carlo Cerrano 4,* , Tamer Özyigit 2,* and Alessandro Marroni 1 1 Research Department, DAN Europe Foundation, 64026 Roseto degli Abruzzi, Italy; [email protected] (S.M.E.); [email protected] (M.P.); [email protected] (A.M.) 2 Department of Computer Engineering, MTF, Galatasaray University, Istanbul 34349, Turkey 3 Cousteau Divers, 75017 Paris, France; [email protected] 4 Department of Life and Environmental Sciences—DiSVA, Università Politecnica delle Marche, 60131 Ancona, Italy * Correspondence: [email protected] (C.C.); [email protected] (T.Ö.); Tel.: +90-542-687-1410 (T.Ö.)

Received: 17 September 2018; Accepted: 29 October 2018; Published: 20 November 2018

Abstract: Dive computers have an important potential for citizen science projects where recreational SCUBA divers can upload the depth temperature profile and the geolocation of the dive to a central database which may provide useful information about the subsurface temperature of the oceans. However, their accuracy may not be adequate and needs to be evaluated. The aim of this study is to assess the accuracy and precision of dive computers and provide guidelines in order to enable their contribution to citizen science projects. Twenty-two dive computers were evaluated during real ocean dives for consistency and scatter in the first phase. In the second phase, the dive computers were immersed in sufficient depth to initiate the dive record inside a precisely controlled sea aquarium while using a calibrated device as a reference. Results indicate that the dive computers do not have the accuracy required for monitoring temperature changes in the oceans, however, they can be used to detect thermoclines if the users follow a specific protocol with specific dive computers. This study enabled the authors to define this protocol based on the results of immersion in two different sea aquarium tanks set to two different temperatures in order to simulate the conditions of a thermocline.

Keywords: diver; underwater; SCUBA; temperature measurement

1. Introduction Temperature data from the oceans are very important to understand the ecosystems and assess the human impact on nature. There is large scale monitoring of sea surface temperature using both remote sensing and in-situ platforms [1–4], but there is a lack of depth resolved temperature profiles for inshore regions, besides the use of ‘animals of opportunity’ [5,6] and human volunteers [7,8] which help to monitor the depths/temperature profile data to address this issue. In fact, Citizen Science (CS) is becoming a powerful tool, bringing together scientists, society and decision makers. Recent reviews show that the last decades have seen a tremendous increase in CS projects and participation, thanks to technological advances amongst other elements. CS projects have also made contributions to science [9–12]. Recreational SCUBA divers are one of the important contributors to marine citizen science. Their key instrument is a dive computer (DC) that they need to use in order to monitor depth, time and in some models the amount of gas remaining in their tank. Dive computers include algorithms to predict the likelihood of decompression sickness and calculate the safe ascent rate for a diver. For that reason, the modern electronic DC can be considered as the most significant advancement in diving

Appl. Sci. 2018, 8, 2315; doi:10.3390/app8112315 www.mdpi.com/journal/applsci Appl. Sci. 2018, 8, x FOR PEER REVIEW 2 of 11

Appl. Sci. 2018, 8, 2315 2 of 11 safe ascent rate for a diver. For that reason, the modern electronic DC can be considered as the most significant advancement in diving since the invention of the Aqualung by J.Y. Cousteau. Modern sincedive computers the invention include of the an Aqualung interface to by allow J.Y. Cousteau. the upload Modern of the log dive of computersthe dive to includea mobile an device interface or a todesktop allow thecomputer. upload Although of the log there of the are dive standards to a mobile on devicethe accuracy or a desktop and precision computer. of dive Although computers there are(EN standards 13319) [13], on one the must accuracy be careful and precision while using of dive their computers output in(EN scientific 13319) study. [13], oneThe mustuse of be this careful data whilewill depend using theiron the output scientific in scientific question study. in mind. The useFor ofexample, this data there will dependmay be scenarios on the scientific where questionan error inof mind.1 °C is For unacceptable, example, there in which may be case scenarios improvements where an need error to of 1be◦ Cmade is unacceptable, in the accuracy in which and caseconsistency improvements of dive computer need to be temperature made in the recordings accuracy and [14]. consistency of dive computer temperature recordingsAzzopardi [14]. and Sayer studied 47 dive computers and concluded that caution should be employedAzzopardi when and using Sayer displayed studied and/or 47 dive recorded computers depth and and concluded temperature that caution data from should dive be computers employed whenin scientific using displayedor forensic and/or studies recorded [15]. Their depth conclusion and temperature on temperature data from is based dive computerson the deviation in scientific of up orto forensic5.1 °C from studies nominal [15]. Their values. conclusion Caution on is temperature recommended is based for the on theuse deviationof dive computers of up to 5.1 as◦C depth from nominalmeasurement values. tools Caution in scientific is recommended study or underwater for the use surveying. of dive computers as depth measurement tools in scientificWright study et al. orreviewed underwater the potential surveying. of SCUBA divers as oceanographic samplers when using theirWright dive computers et al. reviewed to augment the potential aquatic temperatur of SCUBA diverse monitoring as oceanographic [14]. They have samplers also attempted when using to theiruse the dive temperature computers to difference augment aquaticbetween temperature surface and monitoring lowest [14temperatures]. They have recorded also attempted by the to useOperational the temperature Sea Surface difference Temperature between surfaceand Sea and Ice lowest Analysis temperatures database recorded [1] and bythe the conductivity, Operational Seatemperature SurfaceTemperature and density andinstruments Sea Ice Analysis respectively, database to provide [1] and thean idea conductivity, of the degree temperature of error and of densitytemperature instruments recordings respectively, of dive computers. to provide an Their idea findings of the degree agreed of errorwith ofthe temperature results from recordings Azzopardi of diveand Sayer computers. [15]. The Their overall findings outcome agreed indicated with the that results in a range from of Azzopardi test temperatures and Sayer between [15]. The 10 overalland 17 outcome°C, the median indicated variance that in was a range 5.1 °C of (with test temperatures a range of –4.0 between to 1.1 10°C). and Furtherm 17 ◦C, theore, median they concluded variance wasthat 5.1identification◦C (with a rangeof the of depth−4.0 toof 1.1the◦ C).thermocline Furthermore, is likely they concludedto be possible that identificationin the future of with thedepth a more of thestructured thermocline collection is likely of data, to be possiblebut will independ the future on the with strength a more of structured the thermocline collection and of the data, quality but will of dependthe diveon computer. the strength of the thermocline and the quality of the dive computer. The aim of thisthis studystudy isis toto simulatesimulate thethe thermoclinethermocline with a pairpair ofof seasea aquariumaquarium basinsbasins andand determine thethe protocolprotocol toto bebe usedused toto detectdetect thethe thermoclinesthermoclines usingusing thethe divedive computercomputer logs.logs.

2. Methods

2.1. Open Water Dive Prior to controlled controlled experiments experiments in in a asea sea aquarium, aquarium, a SCUBA a SCUBA dive dive is planned is planned to verify to verify if the if dive the divecomputers computers are functioning are functioning properly. properly. The The Bosporus Bosporus Channel Channel dive dive site site was was selected, selected, where where thethe thermoclinethermocline due to to the the counter counter flow ﬂow of of Mediterranean Mediterranean and and Black Black Sea Sea water water is iscommonly commonly observed observed at atvarying varying depths depths and and locations. locations. Gregg Gregg et al. et found al. found the theexchange exchange flow ﬂow between between the Sea the Seaof Marmara of Marmara and andthe Black the Black Sea Seato be to bequasi-steady. quasi-steady. The The dive dive site site was was selected selected as as a apotential potential point point to to observe the temperaturetemperature gradients [[16].16]. A A team of 3 divers made a dive to 49 m for a totaltotal divedive timetime ofof 1616 minmin whilewhile carryingcarrying 2121 divedive computerscomputers attachedattached toto acrylicacrylic rodsrods (Figure(Figure1 ).1).

Figure 1. OpenOpen water tests with real dives (March (21 March 21, 2018).2018).

Appl. Sci. 2018,, 8,, 2315x FOR PEER REVIEW 3 of 11

At the end of the dive, depth, time and temperature profile data was uploaded to a desktop computerAt the and end was of theconverted dive, depth, to CSV time format and temperatureusing Subsurface, profile an data open-source was uploaded software to a released desktop computerunder the andGPLv2 was license converted [17]. to The CSV data format was usingalso uplo Subsurface,aded to anthe open-source Diving Safety software Laboratory released database under the(DSL-DB) GPLv2 using license the [17 Diving]. The dataSafety was Guardian also uploaded portal tofor the future Diving processing, Safety Laboratory both for oceanographic database (DSL-DB) and usingdiving the medicine Diving research. Safety Guardian Divers Alert portal Network for future Europe processing, (DAN) created both for a oceanographicdatabase (DB) with and divinga large medicineamount of research. dive related Divers data, Alert which Network has been Europe collecte (DAN)d since created 1994 a within database the (DB) scope with of the a large Dive amount Safety ofLaboratory dive related (DSL) data, project. which has These been dive collected profiles since we 1994re withincollected the scopefrom ofdifferent the Dive models Safety Laboratoryof diving (DSL)computers project. and Theseconverted dive into profiles the original were collected DAN DB from format different known models as DAN of DL7, diving which computers became and the convertedindustry standard into the originalfor dive DAN profile DB data format exchan knownge, asand DAN many DL7, manufacturers which became are the now industry embedding standard a forDL7 dive compliant profile data exchange,download andoption many in their manufacturers dive computers. are now DSL embedding hosts 70,008 a DL7 dive compliant profiles as data of downloadMay 2017, optionuploaded in their by 5659 dive divers computers. [18]. A DSL special hosts section 70,008 in dive DSL profiles DB was as created of May 2017,for the uploaded analysis byof 5659undersea divers temperature [18]. A special data section for in a DSL future DB wasproject created to foridentify the analysis thermoclines, of undersea especially temperature in datathe forMediterranean a future project Sea. to The identify dive described thermoclines, above especially is the first in dive theMediterranean uploaded for that Sea. purpose. The dive described above is the first dive uploaded for that purpose. 2.2. Sea Aquarium Experiments 2.2. Sea Aquarium Experiments Sea aquariums were selected in order to obtain a controlled environment with a high volume of waterSea where aquariums the temperature were selected is instable order and to obtain will not a controlled be biased environment by the temperature with a high of volume the dive of watercomputers. where At the the temperature experiment is site, stable the and Panama will not (A beQ1) biased and Pacific by the (AQ2) temperature tanks of Istanbul the dive computers.Aquarium At(Istanbul, the experiment Turkey) site,are theequipped Panama with (AQ1) high and prec Pacificision (AQ2) thermostats tanks of Istanbulcoupled Aquariumto heaters (Istanbul,and are ◦ Turkey)stabilized are atequipped 17 °C and with 25.5 high°C, respectively. precision thermostats The tanks coupledwere chosen to heaters to have and su arefficient stabilized depth, at as 17 diveC ◦ andcomputers 25.5 C, need respectively. a minimum The depth tanks to were trigger chosen the depth/temperature to have sufficient depth, profile as logging. dive computers The minimum need adepth minimum is different depth for to trigger each manufa the depth/temperaturecturer but ranges profile from 80 logging. cm to The120 minimumcm. In this depth experiment, is different the forimmersion each manufacturer was at least but1.8 m ranges in order from to 80ensure cm to the 120 activation cm. In this of the experiment, dive computers, the immersion and the proper was at leaststartup 1.8 mof inthe order dive to computers ensure the activationwas checked of the by dive a diver computers, (Figure and 2). the The proper distance startup between of the divethe computersaquariums was about checked 50 bym and a diver at the (Figure end of2). each The immersion distance between the dive the computers aquariums were was carried about inside 50 m anda bucket at the containing end of each water immersion at the stabilized the dive computers temperature were of carriedthe aquarium inside a in bucket order containingto prevent watercooling at theduring stabilized transport. temperature of the aquarium in order to prevent cooling during transport.

Figure 2. Diver checking the proper bootingbooting of the dive computers.

Appl. Sci. 2018, 8, 2315 4 of 11 Appl. Sci. 2018, 8, x FOR PEER REVIEW 4 of 11

The following describes two experiments to simulate the passage of a diver through a thermocline The following describes two experiments to simulate the passage of a diver through a layer while descending and ascending. The ﬁrst phase is designed for stabilization of initial conditions thermocline layer while descending and ascending. The first phase is designed for stabilization of and is not used for modeling the DC response. The stay in each aquarium was extended to one initial conditions and is not used for modeling the DC response. The stay in each aquarium was hour in order to observe, and if possible, model the latency of the dive computers by measuring the extended to one hour in order to observe, and if possible, model the latency of the dive computers by temperature (Figure3a,b). The reference temperature was measured by a laboratory grade instrument measuring the temperature (Figure 3a,b). The reference temperature was measured by a laboratory (556 Handheld Multi parameter, YSI Incorporated, Yellow Springs, OH, USA) calibrated by the grade instrument (556 Handheld Multi parameter, YSI Incorporated, Yellow Springs, OH USA) accredited laboratory of the Turkish Standards Institute. Dive proﬁles were downloaded following the calibrated by the accredited laboratory of the Turkish Standards Institute. Dive profiles were same procedure as the open water dives in the Bosporus Channel. downloaded following the same procedure as the open water dives in the Bosporus Channel.

(a)

(b)

FigureFigure 3.3. ((aa)) Temperature Temperature profiles proﬁles for Experiment 1; ( b) temperature profiles proﬁles for Experiment 2. TheThe orderorder ofof experimentexperiment 11 isis inverted.inverted.

TheThe divedive computerscomputers werewere keptkept forfor oneone hourhour inin AQ1AQ1 andand werewere immersedimmersed inin AQ2AQ2 forfor anotheranother hourhour andand immersedimmersed againagain inin AQ1AQ1 forfor anotheranother 6060 min.min. Periods 1 andand 22 werewere usedused forfor modelingmodeling thethe divedive computers.computers.

3.3. Results DuringDuring thethe openopen waterwater dives,dives, threethree divedive computerscomputers failed:failed: Two hadhad downloaddownload problemsproblems whilewhile oneone shut down down during during the the dive. dive. Hence, Hence, Experiment Experiment 1 in1 the in aquarium the aquarium involved involved 18 computers 18 computers while whileExperiment Experiment 2 involved 2 involved 19, because 19, because the one the that one sh thatut down shut during down duringthe dive the had dive recovered had recovered after a afterfirmware a ﬁrmware update. update. The dive The computers dive computers with download with download problems problems were wereeliminated eliminated from from the theexperiments. experiments. TheThe downloadeddownloaded depthdepth andand temperaturetemperature datadata fromfrom thethe openopen waterwater divesdives inin BosporusBosporus isis givengiven inin FigureFigure4 4.. A A similar similar degree degree of of scatter scatter is is observed observed by by Wright Wright et et al. al. [ 14[14]] and and Azzopardi Azzopardi and and Sayer Sayer [15 [15].].

Appl. Sci. 2018, 8, x FOR PEER REVIEW 5 of 11 Appl. Sci. 2018, 8, 2315 5 of 11 Appl. Sci. 2018, 8, x FOR PEER REVIEW 5 of 11

Figure 4. Temperature and depth profiles for the dives in Bosporus. Solid lines are for depth and Figure 4. Temperature and depth profiles for the dives in Bosporus. Solid lines are for depth and dottedFigure lines4. Temperature are for temperature. and depth profiles for the dives in Bosporus. Solid lines are for depth and dotted lines are for temperature. dotted lines are for temperature. The temperature-timetemperature-time profiles profiles of theof the experiment experiment 1 and 1 2 and are given2 are in given Figures in5 andFigures6 respectively. 5 and 6 respectively.MostThe of the temperature-time DCs Most display of the an DCs exponential profiles display of change anthe expoexperi whichnentialment can bechange1 modeledand which2 are using cangiven the be classical inmodeled Figures equation using5 and thatthe 6 classicalisrespectively. the solution equation Most of a that firstof isthe order the DCs solution system: display of a anfirst expo ordernential system: change which can be modeled using the classical equation that is the solution of a first order system: T (t) = C + A (1 – e−kt) (1) −kt TT (t) (t) = = C C + + A A (1 (1− – ee−kt) )(1) (1) where T is temperature in Celsius, t is time in minutes, 1/k is the time constant (the amount of time it wheretakes for T istemperature temperature values inin Celsius,Celsius, to change tt isis timetime approximatel inin minutes,minutes,y 1/k63.2%1/k is the from time their constant starting (the values amount to their of time final it valuestakes for in temperaturea transient situation) values to of change the system. approximatelyapproximatel C is the yinitial 63.2% temperature from their startingand A is values the step to theirchange finalfinal in temperature.values in a transient Based on situation) this assumption, of the system. the Ccurve is the fitting initial by temperature least squares and method A is the was step applied change to in eachtemperature. DC to find BasedBased the time onon this this constant assumption, assumption, (Table the1). the 1/k curve curve and fitting Rfitting2 values by by least were least squares calculated squares method method for periods was was applied 1, applied 2, 3 to and each to 4 andDCeach to theDC find experimentsto thefind time the constanttime and constant the (Table results (Table1). 1/kare 1). listed and 1/k R and in2 values TaR2ble values were1 for were calculated14 DCs. calculated In for Figure periods for periods 7, 1,a sample 2, 31, and2, 3 curve 4and and 4 fittingtheand experimentsthe graph, experiments where and the theand blue results the line results are was listed aregenerated listed in Table in according 1Ta forble 14 1 DCs.forto measured 14 In DCs. Figure In temperature 7Figure, a sample 7, a values curvesample and fitting curve the redgraph,fitting line graph, where was thewheregenerated blue the line blue according was line generated was to generated temperatures according according to measuredvalues to measuredcalculated temperature temperature using values the and functionvalues the and red (1) linethe is wasdisplayed.red generatedline was generated according toaccording temperatures to temperatures values calculated values using calculated the function using (1) the is displayed. function (1) is displayed.

Figure 5. The results of Experiment 1. Figure 5. The results of Experiment 1. Figure 5. The results of Experiment 1.

Appl. Sci. 2018, 8, 2315 6 of 11 Appl. Sci. 2018, 8, x FOR PEER REVIEW 6 of 11

FigureFigure 6. 6. TheThe results results of of Experiment Experiment 2. 2. Table 1. R2 and time constants for dive computers. Table 1. R2 and time constants for dive computers. Computer Dive 1/k R2 Computer Dive 1/k R2 1 208.7665 0.9947 12 172.7661208.7665 0.70790.9947 Hollis TX1 2 172.7661 0.7079 Hollis TX1 3 165.0955 0.9414 34 241.6872165.0955 0.98870.9414 41 136.4287241.6872 0.99860.9887 12 150.5566136.4287 0.98380.9986 Mares Icon 23 200.0000150.5566 0.69150.9838 Mares Icon 4 143.6142 0.9993 3 200.0000 0.6915 41 143.6142 82.5882 0.99280.9993 2 34.0884 0.9113 Mares Matrix 13 70.923582.5882 0.95140.9928 2 34.0884 0.9113 Mares Matrix 4 51.6683 0.9688 31 76.516870.9235 0.98100.9514 42 24.072051.6683 0.89140.9688 Mares Smart 13 75.545176.5168 0.98320.9810 24 65.060024.0720 0.98770.8914 Mares Smart 31 152.569275.5451 0.99160.9832 2 5000.0000 0.0566 Oceanic OC1 4 65.0600 0.9877 3 153.0933 0.9118 14 149.1615152.5692 0.99530.9916 2 5000.0000 0.0566 Oceanic OC1 1 281.1542 0.9909 32 170.8180153.0933 0.81370.9118 Oceanic Veo 2 No.1 43 205.8048149.1615 0.94440.9953 14 265.7876281.1542 0.98740.9909 21 279.5609170.8180 0.98760.8137 Oceanic Veo 2 No.1 2 95.2336 0.6806 Oceanic Veo 2 No.2 3 205.8048 0.9444 43 213.8564265.7876 0.93660.9874 4 261.9680 0.9879 1 279.5609 0.9876 21 236.362095.2336 0.99150.6806 Oceanic Veo 2 No.2 2 127.9218 0.8740 Oceanic VT4 3 265.2889213.8564 0.95820.9366 4 254.4623261.9680 0.98810.9879 1 194.9532236.3620 0.99980.9915 2 198.2694127.9218 0.99790.8740 SensusOceanic Ultra VT4 No.1 3 213.8182265.2889 0.99980.9582 4 199.9627254.4623 0.99990.9881 1 194.9532 0.9998 2 198.2694 0.9979 Sensus Ultra No.1 3 213.8182 0.9998 4 199.9627 0.9999 Sensus Ultra No.2 1 192.0657 0.9998

Appl. Sci. 2018, 8, 2315 7 of 11

Table 1. Cont. Appl. Sci. 2018, 8, x FOR PEER REVIEW 7 of 11 Computer Dive 1/k R2 2 210.8371 0.9978 1 192.0657 0.9998 3 202.1318 0.9998 2 210.8371 0.9978 Sensus Ultra No.2 34 202.1318203.6378 0.99981.0000 41 203.637871.8781 1.00000.9710 2 93.6115 0.8035 Suunto D4 1 71.8781 0.9710 23 93.611531.0285 0.80350.8399 Suunto D4 34 31.028591.5816 0.83990.9661 4 91.5816 0.9661 1 203.3433 0.8472 Suunto D6 12 203.34337.3099 0.84720.9009 2 7.3099 0.9009 Suunto D6 3 9.8825 0.7992 3 9.8825 0.7992 1 39.4032 0.9654 1 39.4032 0.9654 22 25.680325.6803 0.32200.3220 SuuntoSuunto 33 22.750522.7505 0.04920.0492 44 50.162250.1622 0.96920.9692 11 99.328999.3289 0.95970.9597 22 2.63562.6356 0.09630.0963 UwatecUwatec Galileo Galileo Sol Sol 33 94.105694.1056 0.81010.8101 4 102.1418 0.9663 4 102.1418 0.9663

Figure 7. Sample curve fit ﬁt for the dive computer Mares Smart. The resulting time constant is 66.06 s with an R2 of 0.9877.

Most of the DCs fitfit equation E1 with a highhigh confidenceconfidence (R22 > 0.9 for 41 curves out of 56) while time constantsconstants rangerange fromfrom 7 7 to to 280 280 s. s. Four Four curves curves resulted resulted in ain very a very poor poor fit and fit areand indicated are indicated by italic by italicnumbers numbers in Table in 1Table. As expected, 1. As expected, the time the constants time constants for periods for periods 1–4 and 1 2–3–4 and yielded 2–3 yielded similar results.similar Accordingresults. According to these to results, these ifresults, a diver if staysa diver more stay thans more 3 min than at 3 a min depth at a where depth the where temperature the temperature exhibits exhibitsa stepwise a stepwise change (thermocline), change (thermocline), a noticeable a noticeable change in change temperature in temperature of around of 50% around will be50% observed will be observedfrom the logfrom of the a DC. log The of a resulting DC. The protocolresulting for protocol divers for is described divers is described below: below: Step 1. Mark the geolocation of the dive site. Step 2. Make sure that the DC time and date are set correctly. Step 3. Mark the DC model and serial number Step 4. During the pre-dive briefing, if dive team members are not familiar with the CS project, inform them that you will stop for at least 3 min to mark the thermocline.

Appl. Sci. 2018, 8, 2315 8 of 11

Step 3. Mark the DC model and serial number. Step 4. During the pre-dive brieﬁng, if dive team members are not familiar with the CS project, inform them that you will stop for at least 3 min to mark the thermocline. Step 5. Before the dive, avoid exposing your DC to extreme temperature differentials; if possible, let it be immersed in water at a depth shallower than the one that triggers depth proﬁle logging and decompression computing (30 cm is an adequate depth to immerse for most of the available DCs in the market). Step 6. During the dive, when you will stop for at least 3 min, make sure that this stop does not impose any danger to you and other team members. Such dangers might include but not limited to disadvantageous decompression obligations, drifting with the current or boat accidents. Step 7. Do not limit your contribution with a single dive at the same location. If possible, provide multiple dive data samples for the same location.

4. Discussion The Mediterranean Sea is considered a miniature ocean, able to mimic on a shorter time scale the dynamics of the world’s oceans. This model can be used as a giant mesocosm of the open oceans [19], with various impacts interacting synergistically [20]. Climate change, affecting the usual frequency and intensity of winds, rainfalls and temperature, is rapidly modifying the typical circulation pattern of the Mediterranean Sea. The thermal anomalies documented with an increasing frequency in the last 20 years are showing impressive effects on the distribution of marine organisms. The biodiversity of the Mediterranean Sea is considered incredibly high compared to other areas, owing to the unique and complex geological history of the basin. Owing to the Messinian crisis and the glacial ages, we found organisms with both a cold (northwestern basin) and warm (southeastern basin) affinity. Climate change is creating many problems for species with a cold affinity who cannot adapt to the new thermal regime, and it is enhancing the distribution of warmwater species. Throughout the year the Mediterranean Sea is characterized by a typical thermic cycle, reflecting the parallel distribution of nutrients and chlorophyll. The cycle is characterized by the establishment, during the summer season, of a thermocline that is a variable extended zone defined by a steep temperature gradient. For the marine environment, the thermocline depth is an ecological boundary, because it represents the level of the physiological thermic limit for many species. Nutrients, oxygen, and other limiting factors tend to accumulate near it. Since the global thermal shift, which occurred as a consequence of the El Nino in 1998, the thermocline has been increasing its depth range and many species with a cold-affinity began to disappear from the upper layers in the North–Western basin. Mainly benthic filter feeders are affected by massive mortalities, rapidly changing the seascapes of many coastal areas [21–23]. Even if seawater temperature is only one of the parameters responsible for mortality outbreaks, knowing the precise position of the thermocline during the summer and autumn seasons could allow a better understanding and potentially be used to predict massive mortality events or other important biological processes. The opportunities for researchers to collect this information on a wide spatial scale are very scarce. The contributions of recreational divers equipped with adequate computers could allow for the understanding of irregular patterns of mortality events that, with the poor spatial resolution currently available for seawater temperature close to the seafloor, remain without a clear explanation. The suggested protocol of staying 3 min at the depth of the step change in temperature can detect all the simulated thermoclines of Experiment 1 and 2 but cannot detect the magnitude of the temperature shift. This protocol can be used to locate the depth and the geolocation if used together with dive logging software, such as eDiverlog (Pelagic Pressure Systems, San Leandro, CA, USA). Further expectation on quantitative information of temperature may lead to erroneous data, since a dive computer’s primary aim is not to measure the temperature accurately. For instance, Appl. Sci. 2018, 8, 2315 9 of 11 during Experiment 1, an oscillation of temperature was recorded by a dive computer, probably due to the insufficient digitization. The temperature of 24.15 ◦C is either quantified as 24.4 or 23.9, depending uponAppl. Sci. microclimate 2018, 8, x FOR changesPEER REVIEW and due to poor discretization (Figure8). 9 of 11

Figure 8.8. Temperature oscillationoscillation asas loggedlogged byby aa divedive computercomputer (Hollis(Hollis TX1).TX1).

DiversDivers generallygenerally areare keenkeen citizencitizen scientistsscientists thatthat areare happyhappy toto collectcollect informationinformation toto helphelp betterbetter understandunderstand thethe oceans oceans [24 ,[24,25].25]. This This is confirmed is confir bymed the unprecedentedby the unprecedented success of success DSL in contributionof DSL in tocontribution diving medicine to diving [26 ].medicine The DSL [26]. DB The now DSL has DB a section now has for a detecting section for thermoclines detecting thermoclines and is ready and to acceptis ready data to uploadsaccept data from uploads volunteers; from however, volunteers; a filter however, needs to a be filter added needs to the to webbe added interface to inthe order web tointerface check if in the order divers to followedcheck if thethe correctdivers protocol.followed That the correct will open protocol. a new “the That citizen will open science” a new channel “the forcitizen divers science” who want channel to contribute for divers to who oceanography. want to contribute to oceanography. 5. Conclusions 5. Conclusions Caution should be employed when using displayed and/or recorded depth and temperature Caution should be employed when using displayed and/or recorded depth and temperature data from dive computers in scientific studies. Based on the results, in order to tag the thermoclines, data from dive computers in scientific studies. Based on the results, in order to tag the thermoclines, we suggest that divers stop for a minimum period of 3 min during descent if the diving conditions do we suggest that divers stop for a minimum period of 3 min during descent if the diving conditions not impose any stress including decompression. Our protocol is based on controlled experiments and do not impose any stress including decompression. Our protocol is based on controlled experiments provides a robust method of identifying the tempo-spatial distribution of Mediterranean thermoclines. and provides a robust method of identifying the tempo-spatial distribution of Mediterranean This assessment is critical for environmental impact assessments and vital for initiating timely measures. thermoclines. This assessment is critical for environmental impact assessments and vital for The existence of DSL DB, with 5659 divers already enrolled, offers a promising infrastructure to initiating timely measures. The existence of DSL DB, with 5659 divers already enrolled, offers a launch this marine citizen science action for thermocline identification, enhanced by the Cousteau promising infrastructure to launch this marine citizen science action for thermocline identification, Divers network. enhanced by the Cousteau Divers network. Author Contributions: Conceptualization: S.M.E., P.Y.C. and C.C.; Methodology: S.M.E., C.C. and T.Ö.; Investigation:Author Contributions: S.M.E., P.Y.C., Conceptualization: M.P. and T.Ö.; Data S.M.E., Curation: P.Y.C. M.P. and and C.C.;C.C.; FormalMethodology: Analysis: S.M.E., T.Ö.; Writing—Original C.C., and T.Ö.; Draft:Investigation: S.M.E.; Writing—ReviewS.M.E., P.Y.C., andM.P. Editing: and T.Ö.; S.M.E. Data and T.Ö.;Curation: Project M.P. Administration: and C.C.; S.M.E.Formal and Analysis: A.M., Funding T.Ö.; Acquisition:Writing—Original S.M.E. Draft: and A.M. S.M.E.; Writing—Review and Editing: S.M.E. and T.Ö.; Project Administration: S.M.E. Funding:and A.M.,This Funding study Acquisition: was funded S.M.E. by the and Green A.M. Bubbles RISE project, H2020-MSCA-RISE-2014. The project has receivedFunding: funding This study from was the funded European by Unionthe Green Horizon Bubbles 2020 RISE research project, and H2020-MSCA-RISE-2014. innovation programme under The project the Marie has Sklodowska-Curie grant agreement No. 643712. This document reflects only the authors view. The Research Executivereceived funding Agency from is not the responsible European for Union any useHorizon that may2020be research made ofand the innovation information programme it contains. under the Marie Sklodowska-Curie grant agreement No. 643712. This document reflects only the authors view. The Research Acknowledgments:Executive Agency isAuthors not responsible would likefor toany express use that their may gratitude be made to of Ali the Ethem information Keskin it who contains. shot the underwater pictures. Authors would like to thank to the management of Istanbul Aquarium for their continuous help during theAcknowledgments: experiments. Authors would like to express their gratitude to Ali Ethem Keskin who shot the Conflictsunderwater of Interest:pictures.The Authors authors would declare like no conflictto thank of interest.to the management of Istanbul Aquarium for their continuous help during the experiments.

Conflicts of Interest: The authors declare no conflict of interest.

References

Appl. Sci. 2018, 8, 2315 10 of 11

References

1. Stark, J.D.; Donlon, C.J.; Martin, M.J.; McCulloch, M.E. OSTIA: An operational, high resolution, real time, global sea surface temperature analysis system. In Proceedings of the OCEANS 2007—Europe, Aberdeen, UK, 18–21 June 2007; pp. 1–4. 2. Roemmich, D.; Johnson, G.C.; Riser, S.; Davis, R.; Gilson, J.; Owens, W.B.; Garzoli, S.L.; Schmid, C.; Ignaszewski, M. The Argo Program: Observing the global oceans with profiling floats. Oceanography 2009, 22, 34–43. [CrossRef] 3. Rudnick, D.L.; Cole, S.T. On sampling the ocean using underwater gliders. J. Geophys. Res. 2011, 116. [CrossRef] 4. Kennedy, J.J.; Rayner, N.A.; Smith, R.O.; Parker, D.E.; Saunby, M. Reassessing biases and other uncertainties in sea surface temperature observations measured in situ since 1850: 1. Measurement and sampling uncertainties. J. Geophys. Res. 2011, 116.[CrossRef] 5. Charrassin, J.-B.; Park, Y.-H.; Le Maho, Y.; Bost, C.-A. Fine resolution 3D temperature fields off Kerguelen from instrumented penguins. Deep Sea Res. Part I Oceanogr. Res. Pap. 2004, 51, 2091–2103. [CrossRef] 6. Simmons, S.E.; Tremblay, Y.; Costa, D.P. Pinnipeds as ocean-temperature samplers: Calibrations, validations, and data quality. Limnol. Oceanogr. Methods 2009, 7, 648–656. [CrossRef] 7. Brewin, R.J.W.; de Mora, L.; Jackson, T.; Brewin, T.G.; Shutler, J. On the potential of surfers to monitor environmental indicators in the coastal zone. PLoS ONE 2015, 10, e0127706. [CrossRef][PubMed] 8. Joyce, A.E. The coastal temperature network and ferry route programme: Long-term temperature and salinity observations. Sci. Ser. Data Rep. Cefas Lowestoft 2006, 43, 129. 9. Bonney, R.; Cooper, C.B.; Dickinson, J.; Kelling, S.; Phillips, T.; Rosenberg, K.V.; Shirk, J. Citizen science: A developing tool for expanding science knowledge and scientific literacy. Bioscience 2009, 59, 977–984. [CrossRef] 10. Dickinson, J.L.; Shirk, J.; Bonter, D.; Bonney, R.; Crain, R.L.; Marin, J.; Phillips, T.; Purcell, K. The current state of citizen science as a tool for ecological research and public engagement. Front. Ecol. Environ. 2012, 10, 291–297. [CrossRef] 11. Thiel, M.; Penna-Díaz, M.A.; Luna-Jorquera, G.; Salas, S.; Sellanes, J.; Stotz, W. Citizen scientists and marine research: Volunteer participants, their contributions, and projecIon for the future. Oceanogr. Mar. Biol. Annu. Rev. 2014, 52, 257–314. 12. Collins, K.J.; Baldock, B. Use of diving computers in brittlestar surveys. Underw. Technol. 2007, 27, 115–118. [CrossRef] 13. European Committee for Standardization. EN 13319, Diving Accessories—Depth Gauges and Combined Depth and Time Measuring Devices—Functional and Safety Requirements, Test Methods; European Standard Approved by CEN on 20 January 2000; BSI: London, UK, 2000. 14. Wright, S.; Hull, T.; Sivyer, D.B.; Pearce, D.; Pinnegar, J.K.; Martin, D.J.; Sayer, M.D.J.; Mogg, A.O.M.; Azzopardi, E.; Gontarek, S.; et al. SCUBA divers as oceanographic samplers: The potential of dive computers to augment aquatic temperature monitoring. Sci. Rep. 2016, 6, 30164. [CrossRef][PubMed] 15. Azzopardi, E.; Sayer, M.D.J. A review of the technical specifications of 47 models of diving decompression computer. Underw. Technol. 2010, 29, 63–72. [CrossRef] 16. Gregg, M.C.; Özsoy, E.; Latif, M.A. Quasi-steady exchange flow in the Bosphorus. Geophys. Res. Lett. 1999, 26, 83–86. [CrossRef] 17. Subsurface. Available online: https://subsurface-divelog.org/ (accessed on 10 June 2018). 18. Divers Safety Guardian. Available online: https://diversafetyguardian.org/ (accessed on 10 June 2018). 19. Bethoux, J.P.; Gentili, B.; Morin, P.; Nicolas, E.; Pierre, C.; Ruiz-Pino, D. The Mediterranean Sea: A miniature ocean for climatic and environmental studies and a key for the climatic functioning of the North Atlantic. Progr. Oceanogr. 1999, 44, 131–146. [CrossRef] 20. Micheli, F.; Halpern, B.S.; Walbridge, S.; Ciriaco, S.; Ferretti, F.; Fraschetti, S.; Lewison, R.; Nykjaer, L.; Rosenberg, A.A. Cumulative human impacts on Mediterranean and Black Sea marine ecosystems: Assessing current pressures and opportunities. PLoS ONE 2013, 8, e79889. [CrossRef][PubMed] 21. Cerrano, C.; Bavestrello, G.; Bianchi, C.N.; Cattaneo-Vietti, R.; Bava, S.; Morganti, C.; Morri, C.; Picco, P.; Sara, G.; Schiaparelli, S.; et al. A catastrophic mass-mortality episode of gorgonians and other organisms in the Ligurian Sea (North-western Mediterranean), summer 1999. Ecol. Lett. 2000, 3, 284–293. [CrossRef] Appl. Sci. 2018, 8, 2315 11 of 11

22. Garrabou, J.; Coma, R.; Bensoussan, N.; Bally, M.; Chevaldonné, P.; Cigliano, M.; Diaz, D.; Harmelin, J.G.; Gambi, M.C.; Kersting, D.K.; et al. Mass mortality in Northwestern Mediterranean rocky benthic communities: Effects of the 2003 heat wave. Glob. Chang. Biol. 2009, 15, 1090–1103. [CrossRef] 23. Rivetti, I.; Boero, F.; Fraschetti, S.; Zambianchi, E.; Lionello, P. Anomalies of the upper water column in the Mediterranean Sea. Glob. Planet Chang. 2017, 151, 68–79. [CrossRef] 24. Nguyen, H.Y.; Pedersen, O.; Ikejima, K.; Sunada, K.; Oishi, S. Using Reefcheck monitoring database to develop the coral reef index of biological integrity. J. Fish. Aquat. Sci. 2009, 4, 90–102. [CrossRef] 25. Pikesley, S.K.; Godley, B.J.; Latham, H.; Richardson, P.B.; Robson, L.M.; Solandt, J.; Trundle, C.; Wood, C.; Witt, M.J. Pink sea fans (Eunicella verrucosa) as indicators of the spatial efﬁcacy of Marine Protected Areas in southwest UK coastal waters. Mar. Policy 2016, 64, 38–45. [CrossRef] 26. Cialoni, D.; Pieri, M.; Balestra, C.; Marroni, A. Dive Risk Factors, Gas Bubble Formation, and Decompression Illness in Recreational SCUBA Diving: Analysis of DAN Europe DSL Data Base. Front. Psychol. 2017, 8, 1587. [CrossRef][PubMed]

© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).