375
1 INTRODUCTION
1.1 Context
Occasionallyandforseveralreasons, the navigators
havetoperformsurveyingwork.Insomescenarios,
such as natural disasters, urgent survey operations
needtobe carriedoutinlittleavailabletime, asthe
resultofconstraintsfromtheimperativerequirement
to conduct ship‐to‐shore humanitarian relief
operations.Inothercases,thepresenceofunreliable
chart informat
ion and the need to conduct
amphibious operations in uncharted areas require
rapid action from the fleet in order to deliver the
minimum conditions for the safe passage of ships
towardsrestrictedwaters.
Eventually, it may occur that the danger,
infrastructureorportfacilit
yhasnotbeensurveyed
andthattheshipneedstocollectsufficientdatatobe
forwardedtotheresponsibleHydrographicOffice.
Amongmarinersandothermaritimestakeholders
there is a large consensus about the existence of
advantagestobederivedfromtheimplementationof
Electronic Navigational Charts (ENC) and their
v
isualization in the Electronic Charts Display and
Information System (ECDIS). Simultaneously, those
electronic products have paved the way for the
introduction of several automatic functions which
support or replace a large number of tasks carried
outbythebridgeofficer.Therefore,thereliabilityof
these automatic functions, which has been
incorporat
ed within the Integrated Navigational
System (INS), is extremely dependent on the
existenceofup‐to‐dateENC.
From the user’s perspective, this means that the
presence of unreliable hydrographic information is
no longer a matter of only redefining the safety
margins,asitcanalsoleadtoerroneousperforma
nce
oftheINS,relatednottosomemalfunctioningofthe
INS but to the existence of wrong data serving as
basis for the automatic functions. As Andy Norris
Emergency Survey Toolkit for Naval Operations
V.P.daConceição
EscolaNaval,Lisbon,Portugal
ABSTRACT:Inordertodelivertheminimumsafetyconditionsformovementoftheshipstowardsrestricted
waters, urgent survey operations are required whenever we deal with natural disasters; unreliable chart
informationorunchartedareasrequires.Inrecentyears,therehasbeenahugedevelopmentinpositioningand
surveytechnology.Simult
aneously,chartsproductiontechniquesandGISsoftwareareeasilyaccessible.Inthis
circumstance,aresearchprojectwasmadeinordertoassessthepossibilityofdevelopingexistingcapabilities
ofemergencyhydrographicsurvey.Thetoolkitwasdesignedtoallowtheswiftproductionofusablebottom
representationandsurveyofnavigationalaids,withafocusonnavigationalsafety,rathertha
nbottomcontour
accuracy.
http://www.transnav.eu
the International Journal
on Marine Navigation
and Safety of Sea Transportation
Volume 9
Number 3
September 2015
DOI:10.12716/1001.09.03.10
376
hassaid,theseautomaticsystemswilllikelyallowfor
an ever increasing automation of the planning
process, ensuring that the information is up‐to‐date
andofhighintegrity(Norris.2011a).
In recent years, there has been a huge
development in positioning and surveying
technologies, which provide higher accuracy and
datacollectioncapability(IHO. 2005). Concurrently,
charts production techniques and Geographic
InformationSystems(GIS)haveevolvedsignificantly
(Lee. 2008) (Lam et al. 2007). More recently, the
International Maritime Organization (IMO) has
identified the need to investigate the best way to
automate the collection of internal ship data for
reporting(IMO,
2014).
1.2 ExpediteHydrographicsurvey
The purpose of expedite hydrographic survey is to
supportsafepassage,thereforeitisnotnecessaryto
produceastandardrepresentationoftheseabed.Itis
important,however,toprovideallthedetailsofthe
work and an understanding of the achieved
accuracy. This principle
is determinant for an
assessment of existing risks and for the coming up
withafinaldecisionastowhetherornottoproceed
intothesurveyedarea.
While some manual includes information about
the use of GPS receivers and recreational echo
sounders, there is still plenty of space for further
considerations and evaluations on how to better
exploitGPSreceiversanddigitalechosounders,and
specifically on how to integrate them with GIS
applications. Additionally, some of the techniques
used by professional surveyors, for instance to
analyse and estimate the errors, could be explored,
which might help deliver a work
with a higher
standardofquality.
GIS is one of the most valuable tools presently
available that could offer remarkable advantages at
all stages of the survey work, from the planning to
the presentation of the information. On the bridge,
because of the ECDIS Performance Standards
requirements, it could bring other
advantages.
Despite the possibility to add other overlays in the
ECDIS, such as RADAR, AIS and, meteo‐
oceanographicdata,theirflexibilityandportrayalare
limited and do not allow for the full exploration of
other position‐related information. “What is really
requiredisadisplaythatactsmuchmoreas
ageneric
GIS, using ENC data as its background” (Norris.
2011b).
At some stage, it was expected that the research
would have to cope not only with the ECDIS
limitationsto portray other layers but also with the
navigator’sinabilitytouseandmanageENCdataon
alternative systems. “A fundamental
job of the
human navigator today is to be the integrator of
diversenavigationalinputstoensureconsistencyand
therefore integrity” (Norris. 2011a). Ultimately, in a
nearfuture,navigatorswillhavetoabandonsomeof
the more traditional techniques, applied on paper
charts,anddevelopskillsbasedonGIStools.
1.3 Realcases
In2008,whenthetallshipNRPSagreswasentering
the port of Beira, in Mozambique, the Captain had
the perception that the channels had a different
configurationfromtheonewhichtheofficialnautical
chartpresented.Basedontheexistinginformationhe
decidedtoanchor out of
the port oneday ahead of
schedule and to carry out his own survey with the
support of the Portuguese Hydrographic Institute
(IHPT).Toconductthesurvey,thenavigatorwiththe
help of some element of the ship staff, combined a
low‐cost GPS receiver and a recreational echo
sounder
installed on a RHIB. The raw data was
extractedfromtheequipmentandsentelectronically
totheIHPT.Amongthedepthinformationtheyfixed
and identified some navigational aids. The IHPT
analysed the data and included some degree of
corrections to account for some of the expected
errors. Then, they modelled
a contour layer on the
same projection, datum and scale of the nautical
chart. The information about the navigational aids
had undoubtedly demonstrated that they were
positionedindifferentplaces.
Figure1.Geo‐referencedimagewithadepthcontourlayer
As Rear Adm. Jonathan White, Commander,
Naval Meteorology and Oceanography Command
has said, “Hydrographic surveys are necessary in
order to determine navigational hazards that could
impede the egress of Navy assets involved in the
relief support and enable the flow of humanitarian
supplies”(NMOC,2010).Thatiswhathappenedon
the
island of Madeira, after the storm floods in
February 2010. Immediately after the floods the
Portuguese Navy ordered the IHPT to perform
urgent hydrographic surveys at the port of Funchal
andatotherfloodedplacesalongthecoast.Thefirst
objective of the survey data was to immediately
detect underwater
objects brought by the floods,
which then served to update nautical charts and to
establish the requirements for further hydrographic
surveys. On the debriefing the task leader pointed
outtheimportanceofdevelopinganintegratedand
deployablesurveytoolkit.
1.4 Scopeofthestudy
David Last, as President of the Royal
Institute of
Navigation,haswritten:“Asshipsbecomelargerand
faster, with only one or two to operate the bridge,
and as officers cease to be familiar with traditional
labour intensive, highly‐skilled, visual navigation,
377
canwefindwaysofcollectinganddisplayingallthe
information they need in electronic forms that are
clearandutterlyreliable?”(Last.2008).
Thisresearchaimedtoevaluatethepossibilityof
expanding the existing capability of the Portuguese
navy to carry out emergency hydrographic survey
work.Presentcapabilityshould
bestrengthenedboth
inquantitativeandqualitativeterms.Thismeansthat
this research should provide solutions for the
followingproblems:
1 How to survey a larger area simultaneously, by
providing the operation command with more
optionsforthemovementsofthenavalforces?
2 How to present nautical information with
increasedqualityassessment?
3 How to provide the most appropriate data and
presentation format for a better integration in
availableinformationsystems?
Theconstraintsforthedevelopmentofthesurvey
toolkitstudiedinthisprojectwereassociatedwith:
1 The fact that on‐board staff are not survey
specialists;
2 The
costofthesolutionmustbebalancedwiththe
probabilityofoccurrenceofthesescenarios;
3 Thetime requiredin these situations is the most
importantconstraint.
2 METHODOLOGY
Theproposedmethodologyaimedto:
1 Analysethecurrentoperationalcapabilityandthe
requirementsforthistypeofoperations;
2 Define
evaluationcriteriaforthevalidationofthe
solution;
3 Study a modelled solution for collecting data,
analysing it and producing paper/digital charts.
This should be done by maximizing the
operational use of: Low‐cost handheld GPS
receivers,recreationalechosoundersandGIS.
4 Perform hydrographic surveys and compare the
resultswith
standardhydrographicproducts.
5 Present recommendations and guidelines for the
utilizationofthetestedtoolkit.
2.1 Toolkitdesign
One of the commitments of this research was to
develop a hydrographic survey toolkit using
handheld single frequency receivers, low‐cost
sensors and, if possible, software extant on‐board.
Forbudgetpurposes,
considerationswerebetakenof
therequiredmaintenancecontracts,specificsoftware
and the training of operators. The following factors
wereconsidered:
1 Cost;
2 Accuracy;
3 Integratedsystem:forpositioningandsounding.
4 Training and maintenance: usable by on‐board
personnel,designedforoutdoorusage.
5 GIS data collection software: independent
softwareorequipmentfirmware;
6 Trends;
7 Maximize the applications: explore other
applicationsforthetoolkit.
Based on the above criteria, a Garmin 525s GPS,
with an external antenna and a double frequency
echo sounder, was used for this research. The
decisionontheselectionoftheESRIGISwas
mainly
mandatedbytwofactors:theuseofanimplemented
system within the IHPT and the selection of a
certifiedandstandardproduct.Nospecificsoftware
was selected to capture the data, firstly to avoid
further cost in software acquisition and secondly to
take the advantage of the integrated GPS receiver
and of the echo sounder with an external NMEA
output.AlaptopwasusedtooperatetheGISandto
capture the data from the equipment. For the tide
measurements, a calibrated tide tape was used. It
includes a visual and sound alarm for when the
sensortouchesthewater
surface.
2.2 Experimentsandevaluationcriteria
1 Tideandverticalcontrol:toassessthe qualityof
thetideobservation,atrialduringonetidalcycle
was conducted next to a permanent tide gauge
and a portable tide gauge, used by the
hydrographicteams.
2 Planning:thiswasevaluatedbyassessingthe
time
spent in planning and the use of a varieties of
productssourcesandformatsavailableinplace.
3 Horizontalcontrol:the receiverwastestedatthe
IHPT control point; several observation time
series were used to assess the accuracy of the
receiverwithandwithoutEGNOScorrections.
Thepositions
ofthesoundings were assessed by
comparison of the bathymetry and contour data
againstthepublishedhydrographicdatabasedon
surveysconductedonthesameday.
4 Soundings: two trials have been undertaken to
assess the soundings in comparison with
MultibeamEchoSounder(MBES)surveydata.
5 Plottinganalysis:sometechniques
weretestedto
establish a methodology to detect and correct
errors.
6 Charting: quality checks of parameters such as
conformity, scale, and orientation and position
accuracywereexecuted.
3 RESULTS
3.1 Tide
A visual measurement of one tidal cycle was
conducted next to the permanent radar gauge and
eachobservation
wasbased on theaverageof three
readings.
AfifthorderButterworthfilterwasusedtofilter
the water level data of the radar gauge. The tape
gaugedatawasnotfilteredinthesameway,because
it had a very low sampling rate (one every six
minutes) and only few
observations (85
observations).Instead,amovingaveragefilter,over
fivemeasurements,wasappliedtwice.
Some tests were made to design a suitable and
easily accessible filter; and, in the end, two filters,
withtwostage,wereset.Thefirststageisthesamein
both filters, and consists in a
moving average over
fivemeasurements.Thesecondstage,foronefilteris
378
torunagainthemovingaverage,forthe other is to
use the mathlab filtfilt function. The selection of the
filterwasbasedontheresultsfromthestatistics.The
residuals, shown on the Table I, were computed in
comparison to the water level data of the reference
tidegauge,
whichhasanaccuracyof1millimetre.
Table1. Statistics of the residuals of tape gauge
measurementsandtheestimatedtide(cm)
_______________________________________________
Raw Filter1Filter2 Filter3 Estimated
data (Moving(2x Moving Tide
average)Moving (average(corrected)
average)+filtfilt)
_______________________________________________
Mean 4.13 4.10 4.07 4.14 ‐11.55
Average 3.94 3.57 3.50 3.53 4.34
Deviation
Standard 5.35 4.75 4.61 4.60 4.99
Deviation
Maximum 18.40 15.8 15.09 15.30 ‐20.05
95% 15.57 14.47 14.44 14.53 ‐19.84
Confidence
_______________________________________________
Basedontabledtidedatavalues,atidecurvewas
also computed. Complementarily, a safety margin
canbesettoaccommodatetheerrorsassociatedwith
theverticalaccuracyofthebenchmark.
3.2 Planning
To plan the surveys, several themes were tested
during the two survey trials (Sines and Setúbal),
namely:depth
contourdata, control points, Military
Charts, Nautical charts, Satellite imagery,
Navigationalaidsdata,nauticalpublicationinfo.
Following the compilation of the geo‐referenced
information, it was necessary to set the mission
objective, to collect meteorological and
oceanographic information, and to assess the
availableresources(FIG.2010).Heldontheavailable
informationandinordertoestablishthebestlineof
actionforcrisisresponsesituations,ariskassessment
analysis was designed, based on the following
factors:
1 Firststage:
Positioning method: accuracy, reliability, data
management,autonomy,personnel’sexpertise;
Oceanographic and Weather conditions:
visibility,currents,swell;
Survey boat: manoeuvrability,
attitude,
autonomy;personnel´sexpertise;
Sounding method; accuracy, reliability, data
management, autonomy, tide observation,
personnel´sexpertise;
Areatosurvey:best,recommended,minimum.
2 Second stage, after defining two or three line of
action:
Landpreparationtime;
Processingandplottingtime;
Productappropriateness.
3.3 DigitalDatacollection
All the digital data were captured using the
followingoptions:
1 AhandheldGPSreceiverfirmware;
2 ArcMapGIS;
3 HyperTerminal:tocapturetheNMEAmessages.
3.4 Positioning
Three series of static observation trials were
performed at the IHPT facilities. The control point
wassurveyedbygeodeticGPStechnique,referenced
to
GRS80/ETRS89, with the following quality
parameters:
RMS=16.2mm
Horizontalprecision=0.4mm
Verticalprecision=0.5mm
ThedataretrievedfromtheNMEAmessageswas
used to compute the 2D residuals. A spreadsheets
hadbeendevelopedtoprovidethefollowinggraphic
analysis:
1 Descriptivestatisticsabout
the2Dresiduals;
2 Histogramsaboutthe2Dresiduals;
3 Identificationofthetrackedsatellite;
4 SNRoneachchannelandtheaverageSNR;
5 Diagramcombiningthe2DresidualswithEHPE,
HDOPandPDOP;
6 Diagram combining the 2D residuals with the
numberofsatellitesusedforfix.
One may
verify through the statistics results
(Table2)that,tousethistypeofhandheldreceiver,it
isimperativetoincreasetheoverallprecisionand,at
some level, the accuracy by removing the gross
residuals.Therefore,twoapproacheswereappliedto
remove the larger residuals, one using averaged
solutionsoverdifferent
periodsoftime;andanother
applying filters based on satellite and receiver
parameters.
17.523
4.690
4.623 4.565
4.002
3.622
0
10
20
30
40
50
60
70
Filter 0 Filter 1 Filter 2 Filter 3 Filter 4 Filter 5
0
2
4
6
8
10
12
14
16
18
20
09-Jun
29-Jun
30-Jun
05-A go
06-A go
07-A go
Average
Figure2–Maximumresiduals(m)foreachfilter
The average technique can only be applied for
staticobservation,sinceitisbasedontheassumption
that the GPS receiver is not moving. The second
techniquecanbeappliedinbothstaticanddynamic
modes of operation, although the tuning processes
vary one from another. It is important to note
that
there is no knowledge about the filtering processes
that might exist at pseudorange computation level,
priortothedeliveringofthepositionsolution.
From a practical point of view, since it is not
possible to stay for a long time on the field, about
56%ofthereduction
oftheaverageerrorisobtained
byaveraging15minutesofdata.Forthesecondtype
offiltersthefollowingparameterswereworked:
1 Differentialsolution;
2 Minimumnumberofsatellitesforfixing;
3 MaximumHDOP;
379
4 MaximumEHPE;
5 Velocityandaccelerationover10seconds;
6 Velocityandaccelerationover30seconds.
Thefiltersweresetinaccordancewiththevalues
presented on the Table 3. The filter results were
assessed in relation to variations of maximum
residuals, standard deviation of residuals, mean
residuals and
the number of epochs. Filter 0 was
established to accept only the differential solutions.
Filters1and2weredesignedtoremovethesolutions
basedononly4or5satellites,withhighHDOPand
EHPE,whicharegenerallycorrelatedwithpositions
withlowprecisionandaccuracy.
3.5 Sounding
The
results of the calibration of the echo sounder
(Table4)shownthattheechosoundersystematically
providedlowerdepththantherealvalue,whichwas
consistentwithitspurposeforsafenavigationusage.
Table4.Resultsoftheechosoundercalibration
_______________________________________________
RealResiduals(m)
______________________________________
depth(m) Max Min Mean Standard 95%
deviation
_______________________________________________
1.8 0.90.10.60.19 0.9
_______________________________________________
3.6 Depthreduction
Thereductionofthesoundingshadbeencarriedout
inthefollowingmanner:
1 Integrationoftheobservablesinaspreadsheet;
2 Correction of the positions to account for the
horizontal offset between the transducer vertical
andtheGPSantenna.
3 Computationofthetransducerbeamcone
radius;
4 Computationofdepthuncertainty;
5 Computation of the depth referenced to water
surface;
6 Computationofthereduceddepth.
3.7 Surfacemodel
Once the calculations of the reduced depths were
concluded,thefollowingprocesseswerecarried out
ontheGIS:
1 Creation of a shapefile of points, based
on geo‐
referencedsoundings;
2 ProjectionstoUTM29N,WGS84;
3 Creation of a shapefile of polygons based on
reduced depth (shapefile of points) and a buffer
area defined by the attribute of computed beam
coneradius;
4 CreationofaTINsurface;
5 Computationofsurfacedifference,againsta
TIN
createdwiththeMBESdepths;
Theresultofthesurfacedifferencecalculationisa
polygon shapefile, with the following attributes:
volume, surface area, shape length, shape area and
relativepositioncode(above or below the reference
surface).
A geo‐processing model was built in order to
performtheabove
functionswithasingletool.
Table2.StatisticsoftheGPSstatictrials
__________________________________________________________________________________________________
type Timeseries 95.4%confidence[m] Mean[m] Max[m] Min[m] Standard Variance[m] Count[Nr]
deviation[m]
__________________________________________________________________________________________________
EGNOS 08‐632.2519.99 32.82 4.669.4889.895465
09‐62.131.49 4.01 1.100.300.0910640
29‐621.814.90 69.78 0.257.4555.536715
30‐61.701.39 5.18 0.630.270.0714200
05‐81.610.86 3.67 0.070.460.2114200
06‐8
3.171.32 16.88 0.271.231.5214200
07‐82.181.22 5.63 0.070.690.4814200
SPS 05‐83.612.12 6.60 0.541.011.0214200
06‐87.072.71 27.79 0.172.415.8314200
07‐83.712.09 4.26 0.090.840.7214200
__________________________________________________________________________________________________
Table3.GPSFilterparameters
__________________________________________________________________________________________________
Filter EGNOS Numberof HDOP[m] EHPE[m] Velocity Velocity Acceleration Acceleration
satellites(30s)(ms
‐1
) (10s)(ms
‐1
) (30s)(ms
‐2
) (10s)(ms
‐2
)
__________________________________________________________________________________________________
0 Yes
1 Yes>4<1.8<3.3
2 Yes>5<1.8<3.3
3 Yes>5<1.8<3.3<0.02<0.002
4 Yes>5<1.8<3.3<0.005<0.0005
5 Yes>5<1.8<3.3 <0.002 <0.005 <0.0002<0.0005
__________________________________________________________________________________________________
380
3.8 Plottingandanalysis‐Sinesresults
Generally,thebottomprofileiscorrectlyrepresented
andmostoftheshoalswereidentified.Itisimportant
torememberthatthegoalisnottoproduceanexact
representation of the bathymetry, but to identify a
safe passage for the ship. This means that
it is
necessary to compute a surface that is, ideally,
alwaysabovethe true surface.Naturally, one could
setaverylargesafetymarginbutthismightleadto
the total impossibility of the ship to move ahead
towardstherestrictedwaters.
Considering the corrections and sensors offset
compensations, about
98 % of the resulting bottom
surfacewasbelowthereferencesurface.Alltheareas
below the reference surface were located in zones
withirregularbottomsurfaces,whichdemonstratesa
limitation in the characterization of these types of
surfaces.Inordertoquantifytheoffsetbetweenthe
two surfaces, some analyses
were made, using the
attributes of the surface difference shapefile. The
followingparameterswerecomputed:
1 Percentageofareaaboveandbelowthereference
surface;
2 Volumeaboveandbelowthereferencesurface;
3 Average offset above and below the reference
surface;
4 Averageoffsetforeachshape;
5 Maximum
averageoffset;
The computed results of the first sounding data
set (data set 1) are presented in Table 5, which
consolidate all the results of the Sines trial. At the
next step, the soundings were reduced with an
additionalsafety margin of 1.2m. With this data set
(dataset2),the
resultsshiftedradically,sincearound
70%ofthesurfacewasabovethereferencesurface.
Figure3.Surfaceinterpolationbetweenthesoundinglines
Thenumberofareatermsprovidedanindication
ontheproximityofbothsurfaces,becausethecloser
they were, the larger the number of intersection
zones. Moreover, the averaging effect over the
minimum and maximum values was reduced, as it
was performed over smaller areas, which explains
the appearance of a
maximum offset below the
referencesurfaceof4.673m.Lookingatfigure3,one
can observe that the areas that are below the
referencesurfacefollowapatternassociatedwiththe
soundinglines.Thisdegradationoftheinterpolated
surface should be reduced by using closer spacing
lines and performing the survey
perpendicularly to
the bathymetry. Limitation which resulted from the
factthatthistrialwassimultaneouslyconductedon‐
boardthesurveyvesselperformingaMBESsurvey.
Table5.StatisticsfromtheSinessurveytrial
_______________________________________________
Data Data Data Data
set1 set2 set3set3.1
_______________________________________________
Numberofareas:473 16355 1378 1378
Numberofareasabovethe411 5494 112 112
referencesurface:
Numberofareasbelowthe62 10852 1266 1332
referencesurface:
Numberofareascorrelated‐‐‐34
totheTINlimit
%ofareaabovethereference1.3% 69.8% 95.6%95.6%
surface:
%ofareabelowthereference98.7% 30.2% 4.4% 1.7%
surface:
%ofarearemovedfromthe‐‐‐2.7%
analysis:
Averagedepthoffsetabove0.572m0.264m1.036m1.036m
thereferencesurface
Averagedepthoffsetbelow1.204m0.606m2.464m0.622m
thereferencesurface
Maximumaverageoffset 1.481m0.562m1.036m1.036m
abovereferencesurface:
Maximumaverageoffset 1.204m4.673m6.746m2.696m
belowreferencesurface:
_______________________________________________
Inordertoconsiderthevariationofthesounding
errorinrelationtotheincreaseddepth,anadditional
termwasusedforthesafetymargin.Thisnewterm
was set as percentage of the sounding, and, by
approximation,thevalueof5%wasset.Then,anew
data set of
soundings was computed with a safety
marginof[
depthm
%52.1
].
Furtheranalysishadshownthatalltheareaswith
more than 2.7 metres of offset were related to their
interpolationwiththeouterlimitoftheTINsurface.
Byremovingthoseareasfromthe analysis (dataset
3.1),asignificantreductionofthemaximumaverage
offset was obtained, and yet
2.7 % of the area still
included areas below the reference surface that
needed to be addressed. In order to further clarify
their cause; those areas were graphically correlated
withthecontourbathymetryofthereferencesurface.
As an indication, only 1.3 % of the areas below the
referencesurface
hadanoffsetlargerthan1m,and
most of them were located between the sounding
lines. Although few in number and percentage, for
the purpose of their usage, they are a matter of
concernwhichcanonly be reducedby carrying out
surveys with closer spacing lines, and, considering
that
therewaspreviousknowledgeonthelocationof
irregular bottom surface, by using depth
investigation patterns. For these cases, total
elimination of those areas can only be reached by
carrying out a full coverage survey, for instance a
sweepbarsurvey.
3.9 Plottingandanalysis‐Setúbalresults
The area was
selected in order to test the
methodologyoveranareawithbottomirregularities
andabathymetryprofilerangevaryingfrom4to30
metres.
381
Figure4–Detailofsurfacedifferenceandsoundingofdata
set3,fortheSetúbaltrial(yellowareasremovedfromdata
set3statisticresults).
The results of this survey (Table 6) differs
significantly from the results of the equivalent data
set1oftheSinestrial,since65.1%ofthesurfacewas
abovethereferencesurface,againstthe1.3%ofSines.
This should probably due to the sounding
methodology used at Setúbal, which is much
more
appropriatetotheuseofSBES.
When the safety margin of [
depthm
%52.1 ]
wasapplied,14.7%ofthesurfacewasstillbelowthe
referencesurface,withamaximumaverageoffsetof
1.303 metres. Subsequent adjustment were made to
the variable term, and, when applying 10 %, the
obtainedresultswereimproved,asitcanbeseenon
table6.
Table6.StatisticsfromtheSetúbalsurveytrial
_______________________________________________
Data Data Data Data
set1 set2 set3set3.1
_______________________________________________
Numberofareas:334 391 148 148
Numberofareasabovethe96 61 3 3
referencesurface:
Numberofareasbelowthe235 328 144 132
referencesurface:
Numberofareascorrelated‐‐‐13
totheTINlimit
%ofareaabovethereference65.1% 84.7% 96.2%96.2%
surface:
%ofareabelowthereference34.9% 14.7% 3.8% 2.4%
surface:
%ofarearemovedfromthe:‐ ‐ ‐ 1.3%
analysis
Averagedepthoffsetabove0.776m1.057m1.567m1.567m
thereferencesurface
Averagedepthoffsetbelow0.621m0.446m0.533m0.383m
thereferencesurface
Maximumaverageoffset 0.878m1.058m1.567m
1.567m
abovereferencesurface:
Maximumaverageoffset 1.324m1.303m1.262m1.262m
belowreferencesurface
_______________________________________________
3.10 Charting
The final stage of the survey toolkit was the
production of nautical charts with the relevant
informationcollectedfromthehydrographicsurvey.
OntheGIS,afterthecreationoftheTINsurface,the
surface contours were built with the 3D Analyst
Tools.Inordertocheckthe
propertiesoftheprinted
chart and the GIS layer data, measurements were
madeonboth,andcheckedagainstreadingsfromthe
official nautical chart (INT 1880 – Barra e Porto de
Setúbal). The measurements comprise angles,
bearings, distances and coordinates. The overall
result shown a very close similarity between the
three
formats, with no significant error induced to
theuser.
Figure5.Detailofnauticalchartwithnewcontourlines
4 CONCLUSION
4.1 Conclusions
The information provided by sensors, and by other
external sources, must be considered by the
navigator whenever assessing the safety of
navigation.So,evenknowingthattheymightnotbe
as accurate or precise as the one used by survey
specialists,theymustbeusedasa
complementtothe
informationprovidedbyENC,onapermanentbasis.
GIS presents some considerable advantages in
manipulating geo‐referenced information, and
therefore,itfacilitatesnotonlytheplanningprocess
but also data collection, data analysis and chart
production.
Additionally, a consistent and careful planning
maymitigatealargepart
ofthesourceerrorinthis
type of works, namely by assessing the estimated
parametersofthesatellitenavigationsystem(Seeber.
2003), in conjunction with local environmental
elements, and by implementing procedures for
periodiccalibrationsofallthesensors.
The performance of non‐professionals in the
measurement of water levels, and
the results were
quitesatisfactory,as the residualwas less than 14.5
centimetres(95%).
When conducting the soundings, although the
datacaptureprocessis largelyimprovedbytheuse
of integrated electronic systems, it is essential to
complement the positioning technique with other
independent navigation techniques (leading lines,
hydrographicmarks
orotherpositioningsystem),as
thisistheonlyreliablewaytoreducetheuncertainty
382
ofthepositions,andtodetectblundersinpositions,
inrealtime.
With relatively simple post‐processing
techniques, it was possible to remove most of the
largeresidualsandtoslightlyimprovetheaccuracy
of final GPS/EGNOS solution. Maximum residuals
were reduced from 17.5 metres to 3.6 metres, and
mean
solutionsaccuracywasimprovedbyabouthalf
a metre to 1.3 metres. For static observation, it was
determinedthatthemeasuringshouldbecarriedout
foraperiodof13to15minutes,inordertobalance
theavailabletimeandtheaccuracyimprovement.
The proliferations of firmware and data
formats
are a challenge when considering the integration of
informationfrom different systems. In the maritime
domain the NMEA standard helps to simplify this
process.
The toolkit comprises skills and equipment
available on board Navy ships, with no specific
surveytrainingandaspecialistnavigator.Thetoolkit
alsoincludesaset
ofprocessesandmethodstoplan
andexecuteexpeditiousemergencysurveys,aswell
as to produce survey outputs: a graphical
representationoftheseabottomcoveringtheareaof
interest.
The toolkit was designed to allow the swift
production of usable bottom representation, with a
focus on navigational safety, rather
than bottom
contouraccuracy.
Theresultsdemonstratedthattheyfulfilsomeof
the requirements set by the IHO (IHO, 2008) for
hydrographic surveys, namely for positioning
navigationalaids.However,despitethefactthatthe
IHO requirement of Total Vertical Uncertainty was
notmet,inbothsurveytrials,afterapplyingasafety
margin for the depth reduction, it was possible to
obtain a bottom surface with more than 95% above
the true surface. To further increase the confidence
level, it would be necessary to conduct
complementary survey techniques, namely to carry
outaseepsurvey.
With regard to the current methods used
on‐
board,thismethodologypresentsmuchhigherlevels
ofqualityandapplicability,especiallyasfarasdata
processing and analysis, chart information
compilation and the available format of the final
product―papersordigital.Forthoseprocesses,two
geo‐processing models were created, one for the
assessment of static observation, another
for
processingthereduceddepthintotheGIS.
4.2 Recommendations
Amongst the uses of the EGNOS service, carrier
phase observation with low‐cost receivers could
bringabouttwo major advantages, already used by
professional surveyors, one being the possibility of
measuring attitude data, which then could provide
more accurate corrections
of the soundings, and
additionallyofopeningthedoorsfortheconductions
of surveys out of restricted waters, i.e. where the
swellisexperienced.Thesecondadvantagewouldbe
the possibility of conducting hydrographic survey
withoutperformingwaterlevelobservation.
Another possible avenue for research is the
integration of other types
of sensors available on‐
board,forinstanceimageryfromorganichelicopters
ordatafromAutonomousUnderwaterVehicles.
Finally, as was identified in the course of this
project, ECDIS can no longer continue to portray
additional information, as it might compromise the
safety of navigation. Concurrently, the bridges are
becoming crowded
with displays presenting geo‐
referencedinformation,butfromdifferentsystemsor
sources, the ship itself can collect large amount of
geo‐referenceddatathatneedtobecloselyintegrated
withintheprevailingsystem.Hence,itisnecessaryto
developaGISplatform,whichcanbeflexibleenough
to manage all the
geo‐referenced data available on
the bridge, and sufficiently agile to allow the
navigator to establish different profiles of displays
dependingonthedecisionprocessandtheoperation
in course, without compromising the minimum
standardperformancesapprovedinternationally.
REFERENCES
IHO. 2005. Publication C‐13, Manual on Hydrography. 1
st
Edition (Corrections to February 2011). Monaco,
InternationalHydrographicBureau
IHO. 2008. Special Publication No. 44, IHO Standards for
Hydrographic Surveys. 5th Edition. Monaco,
InternationalHydrographicBureau
IMO.2014.NCSR1/9Developmentofane‐navigationstrategy
implementationplan.London,IMO
FIG. 2010. Guidelines for the Planning, Execution and
Management
of Hydrographic Surveys in Ports and
Harbours. Copenhagen, International Federation of
Surveyors–Commission4
Lam, Steve Y. W.; Leyzack, Andrew E. 2007. Integrating
GIS,ECDISandWeb‐basedMarineInformationSystemfor
Maritime Navigationand Coastal Protection.Hong Kong:
FIG working Week 2007‐Hydrographic Surveying in
Practice
Last, David.
2008. Who are the Navigators? The Journal of
Navigation:RoyalInstituteofNavigation.61,p.1‐6
Lee, Alexander (2008). Marine Information Overlays.
Seaways:October2008.p.8‐11
Naval Meteorology and Oceanography Command
(NMOC). 2010. Naval Oceanography Delivers Capabilities
to Haiti Relief Support. Story Number: NNS100122‐01.
1/22/2010.
http://www.navy.mil/submit/display.asp?story_id=5065
9Accessedon20/12/2010.
Norris,Andy.2011a.Thee‐RevolutioninNavigation.Digital
Ship.March2011,p.42
Norris,Andy.2011b.NavigationandGIS.DigitalShip.May
2011,p.78
Seeber, Günter. 2003. Satellite Geodesy: Foundations,
Methods,andApplications.Berlin[u.a.]:deGruyter
