Home

JR84 Cruise Report - British Oceanographic Data Centre

1. Figure 1 Sea ice concentrations on a 1 March 2003 b 8 March 2003 c 14 March 2003 and d 21 March 2003 Outreach activities Jon Copley Southampton Oceanography Centre Diary webpages A live cruise diary was published on the web as part of the Autosub Under Ice webpages Regular entries were produced aboard the JCR and emailed to Southampton where they were uploaded to the server hosting the AUI webpages The webpages were also made available on the JCR intranet The narrative of the cruise diary presented background the science activities of the cruise through more than 9 000 words of text over 70 images video and audio clips Press contact Press activity was co ordinated through NERC Communications following the Pine Island media plan Press contact passed on to the ship included requests for information from Nature s online news service and images for Newsweek As Autosub was neither lost nor deployed beneath an ice shelf no press releases were prepared Image archive An archive of video footage and still images was compiled during the cruise covering launch and recovery of Autosub ship operations and Antarctic environments including open water sea ice and coastline Over 5 hours of video and 400 still images were shot This material will be used to produce media packages to support future Autosub Under Ice cruises and for outreach activities such as the exhibit at the Royal Society summer sc
2. 0 0 601 133 0 0 209 Mu0 00E9 f 0 OL Sojiul EOnnew aa 0 2 men 1 d pA ejeog puoapuy puesi m jauuey 1oKguinaN 0 07 179 pue s Aeg S 0 0v v9 0 25 yes 1 5 5 0 0 79 5 0 02 9 Keg pue s 0 0 29 0 02 9 134 Rothera Adelaide Island 68 24 0 W 68 12 0 W 68 00 Laubeuf Fjord 67 36 0 S IIE OC Scale aa Kiometers 5 0 Nautical miles 2 0 67 48 0 S 68 24 0 W 68 12 0 W 68 00 135
3. 70 Autosub configuration The scientific payload comprised the following instruments Edgetech 4 12kHz sub bottom profiler Kongsberg Simrad EM2000 200kHz multi beam echosounder configurable to look either up or down WS Envirotech AquaLAB 50 port water sampler Seabird9 CTD with dual CT sensors wetlabs transmissometer fluoremeter and oxygen sensor RDI 150kHz ADCP mounted looking down 300kHz ADCP mounted looking up 71 Trim and Ballast Pete Stevenson Estimates of Autosub s weight volume centre of gravity and volume were made at SOC filename otd1 autosub pete excel AUN AUIantarctic2003 xls dated 5 12 02 where the build was estimated to 10kg buoyant for a water density of 1026kg m Measurements in air and fresh water were made before shipping 6 12 03 and recorded on a Mathcad programme filename TrimBuoyAntarctic2003 mcd dated 06 012 02 Additional ballast of 2 3 kg in the nose and 4 5kg in the tail lead weights was added at SOC to give a final predicted buoyancy of 8 5kg in a density of 1026kg m3 A basic floatation test was carried out in calm waters with the winch lines still attached before any missions were run to ensure the vehicle floated Table 11 shows the changes made throughout the campaign Inspection of the data after mission 312 showed the rate ascent to be practically zero at around 40m depth while floating up without any propulsive power A plot of the water density from the CTD showed a
4. buoys transmit position air temperature and atmospheric pressure every three hours via the Argos satellite system They were unpacked and set up on the after deck two days prior to deployment to check that the Argos transmitters were working The aim was to deploy the drifters in a broad triangle with at least one of them over the continental shelf In the end the positions were dictated by the ice conditions which meant that there was only a narrow latitudinal band between ice that was too loose to risk deployment and ice that was too heavy to penetrate with the ship The final arrangement ended up close to linear and a period of strong southerly winds shortly after deployment pushed all the drifters well north of the continental slope Deployment sites and drift tracks over the main work period of JR84 are shown in Figure 6 72S Thurston Island Figure 6 Sea ice drifter tracks from 18 00 on 11 03 03 to 21 00 on 27 03 03 for Argos buoys 8058 blue 8059 red and 8064 green Original deployment sites are indicated by the stars The 1000 m contour is inaccurate but provides a rough indication of the location of the continental slope 29 Physical Oceanography Mark Brandon Ziggy Pozzi Walker Dan Hayes Adrian Jenkins Open University and British Antarctic Survey The main aim of the physical oceanography programme was to investigate the processes by which Upper Circumpolar Deep Water uCDW intrudes onto the c
5. After examining the data it was decided to change instrument settings to give beams at equal angles rather than equidistant across the seabed Unfortunately the supplied control software did not work satisfactorily and the initial alternative method adding the parameters to the install file suggested by Simrad also did not work mission 313 The next day Simrad supplied a small software utility which allowed us to successfully send runtime parameters to the instrument This was used to set the instrument up for mission 314 The results were still not satisfactory The minimum depth setting was then changed from 3m to 0m with a slight improvement on mission 317 The system did not appear to allow a negative minimum depth setting to be set so it was decided to use a false depth sensor offset of 5 0metres to increase the usable range of depths This finally appeared to give satisfactory results during missions 318 and 321 Mission 319 gave little good data probably due to the high pitch angles encountered during the mission although the parts of the mission where the surface was in range appear to show some sensible data No navigation data was recorded during mission 320 but from looking at the raw data there appears to be valid swath data Unfortunately the Simrad processing 76 software rejected this data It may be possible to retrieve the swath data using alternative processing software although much of the mission was too deep fo
6. 0 Reads the oceanlogger data streams into a PSTAR format and merges in relative wind speed and direction from the anemometer data stream Output files are 84ocl jday a p raw and ocl841 The former of these is the 12 hour data segment for morning a or afternoon p of Julian day jday The latter is the master file to which successive 12 hour sections are appended 84 1 1 Divides the data into ocean data and meteorological data files writing meteorological data to a separate file Output file is 84met jday a p raw containing the meteorological data twvelexec Merges the met data file with gyrocompass and navigation data streams in order to calculate ship motion and true wind velocity Output file is 84met jday a p true Problems Our passage and out of the ice meant that there were frequent periods where the system did not function due to ice blockage of the intake pumps The salinity data are will have to be re calibrated on the basis of the post season check 50 ADCP Measurements Sarah Hardy and Mark Brandon Open University Summary This report describes the method of acquisition of ADCP data on JR84 The system was operated in two modes water track mode when water depths were greater than 500m and bottom track mode in shallower waters In general the ADCP worked very well with water track velocity information generally obtained to 350m depth and bottom track velocity information to 550m T
7. 10 Stbd panel minor damage 11 Bottom panel badly damaged 12 Frame work badly damaged in tapered section lesser damage on parallel section apart from lower port junction with tapered section This is possibly where propeller impact occurred to be assessed 13 Seabird connectorsn damaged by being forced backwards into domes 14 Oxygen and C2 cables damaged 15 SeaPam damaged internally broken ferrites 89 18 03 03 13 53 8 03 03 13 54 90 19 03 03 18 50 20 03 03 13 16 91 Summary of problems encountered during cruise Nick Millard Date mission info Faults symptons Findings actions 03 03 03 between Brabant and Anvers Islands in calm water for buoyancy test and emergency beacon test No problems all looked OK Mission 307 03 03 03 1 TP II system not working 1 None at this time Gerlache Strait S 64 32 5 W 062 29 7 WP2 8 64 32 1 W 062 28 7 Drop weight dive Max depth 50m Mission 308 03 03 03 1 Cable tray hanging loose on recovery 1 Adjusted support to prevent movement of tray Gerlache Strait 2 system not working 2 Removed TP II from fish Drop Weight Dive 3 Wanted to use spare data telemetry transducer new out of box but found Position mode to wp at wrong gender connector fitted Made up adaptor cruise depth of 50m 4 TP II to transducer lead found to be pin to pin spare lead provided by Octopus
8. 900 m M317 8 34 7584 34 7092 1200 m M317 8 34 7492 34 7134 700 m M317 8 33 8797 34 4905 mixed layer halocline M317 8 33 5417 33 4211 Om M317 8 33 4886 33 7308 0 100 m M317 8 33 5062 32 9287 0 10 m M317 8 33 4676 21 5205 Om There appears to be a gap in the CTD file which could offset the 15 minute averaging interval from the actual sampling interval Average of salinity calculated from probes one and two Acoustic Doppler Current Profilers The ADCP collected reasonable data however not over the range it should have An example from mission m323 on 24 March is shown It is clear that the vehicle was changing course as the various components change sign An absolute velocity reference was not available for any run except for m320 21 March which was bottom tracking Unfortunately it was just traveling in a circle at this time On this run the number of 126 bins each 8 m containing valid data from the downward looking ADCP was increased to 11 or 85 m range For all other runs only 4 5 bins 31 40 m were useful in the downward direction while 7 9 bins 55 70 m are typically useful in the upward direction This problem has not been addressed satisfactorily as of yet Previous Autosub data are more like m320 It seems likely that unsuccessful attempts to bottom track are somehow contaminating the downward A
9. MC data logger 94 Mission 315 15 03 03 Same as 314 Turned at southerly point but never made it back to WP Surfaced 800m short of WP was seen to be down at 250m No digital acoustics No Radio link Event 9 mission 314 involved splicing power lead to replace damaged connector this failed Mission 316 16 03 03 Same again Checked abort status 3 3 0 0 unusual set found to be finger trouble Rest of mission seemed to go ok except telemetered depth at 49 meters when on surface should have know what that meant 1 Data had major drop outs descending 600m and ascending 200 0m Sub came aboard not working first time ever 3 Network connector on MC data logger could be manipulated to stop start vehicle 4 Added extension lead to try to ascertain whether bulkhead or harness fault mostly went away 5 Cut into leg to short network leads and then blank leg 6 Cleaned all network connectors 7 Left running on shore power over night lost power and would not power up in the morning suspect switch transistor 8 NOSE OFF 9 Removed power node found dry joint on surge protector fairly convincing fault Mission 317 17 03 03 Down to 1320 metres 1 Completed mission successfully 1 A few dropouts at 200m ish when ascending none descending Mission 318 17 03 03 Run at 100 metres for swath 1 Completed mission successfully 1 No dropouts 95
10. Mission 319 18 03 03 Tests to 150m x 2 and 250m x 2 Completed mission successfully No Seapam transmissions Damaged under counter on recovery Launch was carried out at 1615 in 25 knt SW wind in sea state 4 5 A dive weight was used to quickly take the vehicle below the surface to avoid possible collision with a sprinkling of various size bits of ice The sub surfaced at about 1732 by which time the wind had increased to 35 to 40 knts in a sea state which had increased to 5 6 The jack in the box was fired and a close pass revealed that the recovery line was streamed nicely An attempt to back the ship up to recover the line was aborted because of the strong wind and the ship made a Williamson turn to approach from down wind The second pass was a little too distant for the grapneling party to reach but the third pass was successful The thin line could not be held initially and was made fast until the ship could drop back to relieve the tension In the meantime a lump of ice 2 thirds the length of Autosub drifted into and became entangled in the lines but luckily freed itself and the main recovery lines were attached in the normal way to the line leading aft When passing close to the counter Autosub appeared to accelerate towards the ship and disappeared briefly under the counter and it became apparent that Autosub had sustained some damage Recovery was completed during which the forward recovery line was badly damaged when
11. ctdoff This program requires the file CCctdNNN cond and produces the mean conductivity residual and the standard deviation These numbers should be written on the plot produced from seactd2 for further reference On the basis of the results of seactdl and ctdoff it must be decided whether some bottles should be rejected and the conductivity residual recalculated For example on JR84 the cruise protocol meant that some bottles were fired at depths where the salinity gradient was very steep so that some bottle samples were unsuitable to use in calibration The cond file should then be moved to directory samples cond 84seactd3 This exec requires the output of ctdoff the conductivity residual The conductivity offset is added to the rough version of the ctd file CCctdNNN the output of seactdO and the salinity re derived with this new conductivity The output 15 CCctdNNN cal The file CCetdNNN cal should then be moved to the directory cal 84seactd4 Use mlist to select the downcast It is important to remember that the cast will go down to 10m and return to the surface before starting the true downcast The output files are CCctdNNN 24hz and CCctdNNN 2db These should be moved to the directories 24hz and 2db respectively 84seactd5 This exec similar to 84seactdl uses the updated values of salinity rather than the raw data At this stage the second conductivity and temperature variables are dropped they may be useful
12. Reads Trimble data into pstar format Steps datapup transfers the data from RVS binary files to pstar binary files pcopya resets the raw data flag on the binary file pheadr sets up the header and data name of the file datpik removes data with a dilution of precision hdop greater than 5 Output files 84gps raw just before editing stage 84gps following datpick Ashtec GLONASS GG24 The Ashtec GG24 accepts data from both American GPS and Russian GLONASS satellite clusters giving a constellation of 48 available satellites and should theoretically be more accurate However experiments on previous cruises have suggested that the accuracy is significantly lower than the differential GPS Data were logged routinely using ggexec0 called from jr84 go but were not used in the processing of other data streams Output files 84glo raw 84glo following basic quality control of raw data Ashtec ADU 2 The Ashtec ADU 2 GPS is used to correct errors in the gyrocompass heading that are input to the ADCP The configuration of the receiver is complex made more so by the fact that the receiver can only be configured with the use of a laptop running a terminal emulation program Configuration data for the Ashtec aerial configuration is shown in Table 8 The port aft antenna is designated number 1 port fwd is number 2 stbd fwd is number 3 and stbd aft is number 4 the XYZ vectors have been adjusted so that heading
13. The garage container was a modified open top container fitted with a lifting roof to accommodate the head of the gantry It was also insulated although to a lesser degree none on the floor an omission that needs to be rectified for the future and curtains at the Autosub access end The two were positioned in an offset T formation forward of the gantry so that the submarine could loaded into its garage using the gantry s ability to run its beam inboard Twist locks fitted to adaptor plates designed to fit the standard 1 metre deck matrix were used to fit them to the deck Strops were added as a precaution Heating 12kW lighting and electrical outlets were supplied with both dirty for heating and lighting and clean electrical supplies from outlets on the after deck Telephone fire alarm and public address speaker were supplied and fitted by ship side Mobilisation went well with only minor adjustments needed to align heights and container deck plates To achieve vertical alignment of Autosub and gantry head with the floor and roof of the garage the gantry was mounted on eight 200mm cotton reels A few minor problems were encountered during mobilisation and during operations it became apparent that a few changes could be made to improve the facility e g better defences against rouge waves at the aft end However overall they provided a very satisfactory environment for garaging and working on Autosub in temperatures below 10
14. i e operable in an autonomous environment I note that when using the shipborne EM120 system operator intervention is sometimes required to force the system to lock on to the true bed echo Personally if I am involved with using data from the EM2000 in future I will endeavour to import the raw data into an alternative and more explicit software package such as MB system which is free is installed on the JCR unix system and for which the code is available James Perret believes that decoding the raw files would be a simple task and may proceed with that Mission data assessments After the initial unsuccessful testing in Gerlache Strait with the EM2000 in downward looking mode the system was turned into the upward looking configuration for all of the remaining missions It was deployed in both open water for purposes of calibration and beneath sea ice to collect data on sea ice draft in support of the Brandon project Successful data was collected from both these targets during missions M318 and M321 respectively These data are reproduced below These datasets will be the subject of further analyses by Chris Banks OU as part of the Brandon proposal 110 3687000 000L89 Mean dept 6 Planning Normal Line Active Line Polygon Figure 30 Autosub EM2000 data from Mission M318 showing strong sea surface return 3712500 OOSZTLE 3713000 OOOETLE un nination B Below Plannin
15. m cm s cm s cm s cm s db cm s cm s cm s cm s db degrees 84bot time heading bottomew bottomns depth temp 84bot t time heading bottomew bottomns depth temp OORUN 84bot corr time heading bottomew bottomns depth temp 84bot true time heading bottomew bottomns depth temp a ghdg NOOR WD degrees cm s cm s m degrees C degrees cm s cm s m degrees C degrees cm s cm s m degrees C degrees cm s cm s m degrees C degrees 57 84adp cal 84bot cal 1 Time 1 time 2 bindepth 2 heading 3 Evelcal cm s 3 ebotcal 4 Nvelcal cm s 4 nbotcal 5 Velew cm s 5 bottomew 6 Velns cm s 6 bottomns 7 Melvert cm s 7 depth 8 Velerr cm s 8 temp 9 Ampl db 9 a ghdg 10 Good 11 a ghdg degrees 84adp abs 84bot abs 1 Time S 1 time 2 Lat 2 heading 3 Lon 3 ebotcal 4 Distrun km 4 nbotcal 5 bindepth m 5 J bottomew 6 Evelcal cm s 6 bottomns 7 Nvelcal cm s 7 depth 8 Absve cm s 8 temp 9 Absvn cm s 9 a ghdg 10 Velvert cm s 10 ve 11 Velerr cm s 11 vn 12 Ampl db 12 lat 13 Good 96 13 lon 14 a ghdg degrees 15 Ve cm s 16 Vn cm s 84bot abs 2 Velocity amplitude correction Heading misalignment correction Mean ve Standard deviation of ve Mean vn Standard deviation of vn Mean heading Standard deviation of heading 7
16. system The SBE35 data files were transferred to the directory pstar data ctd ascii_files sbe35 3 Further processing of the CTD data in UNIX Salinities Salinity data from the bottle samples is needed for further processing Using the spreadsheet created with values obtained from the salinometer and with reference to the original deck log samples should be matched up to individual bottles A new spreadsheet file should then be created to contain three variables bottle number botsala and botsalb Missing data should be designated with 999 0 This file should be saved as an ascii file with the filename 84samNNN txt It should then be ftp ed to jruf and placed in the directory pstar data ctd samples SBE35 temperature data There is one file for each day on jruf in the form jday txt This file must be split into 12 records for each station one level for each bottle The file can be created using an editor such as emacs or vi and will again need reference back to the original CTD deck log There is no processing other than the deletion of all records except the 12 relating to the relevant station The data must be saved as a file called CCsbeNNN in the same directory i e pstar data ctd ascii_files sbe35 CTD processing using pstar execs The execs assume that the files are tidied up after each one is run They will check for the files when running and say where the files should be 84seactd0 This exec converts data from seab
17. 0015 20030324 211808 57036 080 100 47 680 570 37 084 100 47 926 1803 200133 1663 30 76 00140008000 015 20030324 194617 RAW S70 37 353 1 100247 005 570 37 351 100246 713 1798 199578 40374 285 72 174 5 20 23 4 015 20030324 191541 RAW S70 37 700 100 46 650 570 37 352 100 47 005 1796 199356 43181 290 07 bo 9 21 66 0015 20030324 184504 57037 178 0 100 46 516 S70 37 700 100 46 650 1799 199689 26120 271 98 002 5 0015 20030324 181427 570 35 713 100 45 879 570 37 176 100 46 516 1787 198357 2351 0015 20030324 171314 RAW S70 32 861 100 43 132 570 34 311 100 44 531 1798 199578 2075 31 49 o 40 ho 0015 20030324 164237 RAW S70 32 121 W100 42 450 S70 32 860 _ W100 43 132 1805 200355 42658 141 9 002 72 33 21 29 0015 _20030324_174350_RAW S70 34 312 W100 44 531 57035 713 100 45 879 1792 198912 1665 34 48 8 Number of times pulse read Number of times pulse read multiplied by number of beams used 119 Summary of Performance of Mission 324 Line Start Posn south Start Posn West End Posn South Posn West No of No Max Depth Min Mean Valid pings soundings m Depth Depth soundings m m 541 8 oof 4238 256 0012 20030325 184408 RAW 570 32 139 W98 30 067 570 32 672 W98 30 296 1807 200577 7536 46 35 0 02 11 84 ___ 376 0012 20030325 191445 RAW _ S70 32 672 W98 30 296 570 32 079 W98 30 544
18. 44 40 Meters e Figure 26 Track plot of first segment for mission M309 Second segment Similarly for segment 2 Now use Matlab to apply clock correction and interpolate depth Note change in time correction to 3 sec Output from SGYreader Opened file data0001180 jsf data0001180 jsf 2003 3 4 17 30 1 999992e 000 Incorrect read of header data0001180 jsf 2003 3 4 17 40 4 999992e 000 recordnum 302 Closed file Opened file data0001181 jsf data0001181 jsf 2003 3 4 17 40 6 Incorrect number for bytesToFollow 1256661583 data0001181 jsf 2003 3 4 17 49 5 999999e 001 recordnum 298 Closed file Sgyreader ended Note that times are now 2 5 seconds on previous read and add depth correction using Depthinterp m and M309 bnv 105 1000 AB e e e sample number 6000 7000 100 200 300 400 500 600 trace number Figure 27 Segment 2 of profiler data from M309 Here the depth of the start of the primary reflection is 3400 samples 432 07 metres 595 5 m and 1000 samples on the plot is 48 07 metres so track at 150 metres above seabed We note that Autosub has tracked the bottom with some precision we could do some calculation on this 106 11700 11900 12100 12300 12500 12700 400 450 500 550 600 650 700 750 Figure 28 Comparable segment showing Autosub track top and bathymetry from shipborne EM120 syst
19. 9 I aw oot Nort s we XN 26 N We Ju N e 47 UN Ld y x 4 A VUE e ev _ M uO 00 LL M 0 0 0LL M 0 00SLL Mee cen i gt er L uie je3ueunuo Aeg pue s M 0 000LL 0 0911 0 08 14 131 M 0 0 LOL 0 0201 0 02601 0 0 oz mem ajeos S 0 0 04 0 0 0L gt an ad A ERR REDE Tope 769 jeyueunuo Aeg t ptt z 0 0101 0 0201 0 0 01 132 M007 086 00 901 5 00 14 at o m ree i m 00 901 5 0 0 04 v y x amp 1 j x as 0 02 86 uie je3usurn uo Aeg pue s 80 L gt i T ae Tim Be ho cd 0 06 2801 tM n 1 i 3 M 0 0 601L Pu d i de n Ho J AT TM r y i d i A f 8 A 6 l NOS a L lt Y N t 4 d
20. A full ocean depth CTD cast station 003 was run At midday local time the ship headed further west to access Pine Island Bay along the W106 meridian 11 Mar 2003 The ship s track was pushed further west because the sea ice extending west from Thurston Island was blocking the intended route We begin heading south along W111 12 Mar 2003 Our southerly progress was halted overnight when we encountered sea ice and icebergs At first light we continued southward along a narrow corridor between the sea ice to the east and a line of icebergs coming from Thwaites Glacier to the west New ice was forming all around the ship The ship turned north through the heavier ice to test the condition of the tongue that could potentially block off our retreat from the Bay The ship ran a swath survey outside the ice overnight 13 Mar 2003 Another Autosub deployment M313 ended with similar problems We started a five station 004 008 CTD section perpendicular to the continental slope 14 Mar 2003 The CTD section was completed overnight and the ship returned to deep water for another Autosub test in the morning M314 It was decided that the continuing freeze up to the south and the westward movement of the ice from Thurston Island put operations in Pine Island Bay out of the question We moved westward to W110 overnight broadening the swath survey along the continental slope 15 Mar 2003 At first light the ship moved out to deep water to start anot
21. Autosub collided with the ship and left the FS AU transducers unserviceable Since only a few short parts of these missions were conducted in bottom tracking mode useful data were acquired over only a short track These data do however indicate that the system is satisfactorily installed and will provide a valuable method of observing the seabed There follows a processing log for the data collected during M309 This processing scheme is not intended to indicate how one dataset was processed to yield profiles corrected for clock drift and vehicle depth Outstanding issues 1 The navigation data is not currently passed to FS AU and so the output files do not contain position information The facility exists but it is arguable whether supplying the raw position data to the FS AU would be of much benefit The final processing needs to take account of the best navigation data and since best navigation data is only available after the mission is completed a level of post processing will always be required 2 There is a question over what is the best way to process the data from the FS AU The TOPAS software used to process the shipborne sub bottom profiler data may be useful but at present there is no clear method for reformatting the jsf data into TOPAS format Edgetech do have software to do this but this costs an additional 7k and we are disinclined to recommend spending extra money on a facility that arguably should have been made av
22. DATAOO077 81 jsf Plot from JSTAR capture function with depth scale by jstar No correction of Vehicle depth or along track position r Transmitter puls e Uu Communications bursts TIT Data gap cf 25 minutes here i Depth nominal metres Nominal distance along track Figure 24 Jstar captured data from M309 and then tarted up in Coreldraw showing the main features of the record Matlab processing Since Jstar is not designed for processing data from an underwater vehicle it seems unlikely that it will prove to be sufficient to process the Edgetech data I have written Matlab scripts that allow further processing but these are not intended as a turnkey system and should be understood and modified as required before they re implemented They are Segyreader m sequentially opens a series of files and calls getsegy to read them Getsegy m reads the jsf format files and calls Unpackhar if required Unpackhdr m reads the header fields if required Depthinterp m reads bnv file and interpolates the vehicle depth for each Edgetech trace and adds this to the start of each trace Imagedisp m displays concentenated files with or without depth Checking for data gaps The first step is to run Segyreader to read files and print out the internal time for the first and last message in each file This allows a determination of whether there are gaps between the files Sample output Ope
23. Day of Year Number 64 65 oo 66 67 TU 68 69 70 71 72 73 s 74 75 76 77 78 79 80 81 2 83 84 85 86 87 88 89 90 91 92 mass 3 ams 129 moam MARI MAO MAAR MAAR MOAK ARO MARE MAO UND MABEL ANOLON Masi na i er r in sa cce setae an Lot a p i 1 ee jee a 1 BoE pf eee Manes m zl esin1 vgur LLL A T ee M M E O 42 I 7 AO MAS ROD MODE MOTE ROR MOORE GM 130 Mu0 00ELL 0 OL ames 0 oc 25 jeonney f d P er Pannen 2 0 dM EL Tp 5 0 0814 AE
24. Global Change and the Antarctic GLOCHANT program Hobart Australia 27 deviation of 39 2 20 3cm n 114 The floe was occupied from 1730Z until 18407 Longitude was 102 00 0 and latitude was 70 51 35 FS3 on the afternoon of 9 March used the ship s ice anchor drill for the first two metres The OU drill head and extension rods were then used attached to the ship s anchor drill for further deeper drilling Three holes were drilled the last of which failed to clear the thickness The depths at the three holes were 2 43m 2 66m and greater than 2 7m excluding snow depth The snow depths at the first two points were 27cm and 34cm There was no measurement made of snow at the location of the third hole as no measure of ice thickness could actually be made either It is perhaps not surprising that the third hole was so deep as the measurement was made adjacent to a surface ridge Snow depth measurements were made on this floe too with a range of between 2cm and 86cm with a mean and standard deviation of 45 3 15 6cm n 107 The floe was occupied from 2015Z until 2118Z Longitude was 102 21 4 and latitude was 70 53 0 28 Sea Ice Drifters Adrian Jenkins British Antarctic Survey Three sea ice drifters supplied by Hartmut Hellmer of the Alfred Wegener Institute were deployed on multi year floes FS1 3 north of Thurston Island This work was carried out on 8 9 March as part of the floe sampling work described above
25. PA200 20 5 2130 26993 not known Altimeter Primary Seabird SBE 43 0245 27 Aug 02 Oxygen sensor Secondary Seabird SBE 43 0242 27 Aug 02 Oxygen sensor Fluorometer Chelsea Aqua 088216 11 june 01 3 All calibration coefficients are given in the Appendix The CTD was connected to an SBE 32 12 position carousel water sampler S N 3215759 0173 carrying 12 10 L bottles In addition the CTD was connected to an SBE 35 Reference Temperature Sensor S N 0315759 0005 33 Deployment of the CTD package was from the mid ships gantry and A frame on a single conductor torque balanced cable connected to the CTD through the BAS conducting swivel This CTD cable was made by Rochester Cables and was hauled on the 10T traction winch There were no problems deploying the CTD package and no re terminations were required throughout the cruise The CTD data were logged via an SBE 11 plus deck unit to a 486 Viglen PC running Seasave Win32 version 5 25 Sea Bird Electronics Inc This new software is a great leap forward compared with the DOS version in that one can draw several graphs of various recorded parameters in real time as well as having numerical lists of data to the screen The data rate of recorded data for the CTD was 24 Hz A full station list is given in table 2 below Calibration of the CTD data Four files were created when the Seasave Win32 version 5 25 module was exited at the end of each CTD cast a binary data file with the extensio
26. Re spliced adapter lead shorting network leads to make SB a stub MISSION 312 10 03 03 1 Reached 1300 m then min EM signal 1 Looked at data revealed network failure similar to before and at times Basic Depth Pressure 2 Acoustic telemetry indicated firing of complete failure Test jack beacon abort dive weight 2 Removed Pressure 2 bottle forward and tested bulkhead connector for Investigations continuing 3 Very slow ascent after prop stopped recovery continuity and shorts to shell etc nothing found on into 11 03 03 line wrapped around it 3 Checked internal connectors and connections in P2 no problem found Spiral down to 1320m 4 Replaced IE 55 19 way bulkhead on P2 despite results from 2 5 Leg 3 shorting connector replaced 6 Looked at cct diag for power node to check effect of plugging external power into network connector power went to 24V and power went to 24V 7 Measured resistances into network value into data logger node on the high side 8 REMOVED NOSE AND TAIL 9 Removed dome 1 front Edgetech to measure resistance into network Problem apparently at back 10 Removed dome 1 rear data logger where resistance is between 3 6 and 2 3 ohms 11 HE 14 connectors suspect change them in P2 data logger 12 Replaced CAT 5 cable to transceiver in data logger and lonworks module 13 Removed 7 5kg ballast 93 MISSION 313 13 03 03 Spiral down to 1320m 1 EM 2000 playing up 2 Network problems not
27. Sediments and meltwater would have been delivered from the ice front directly at the shelf break leading to increased sedimentation rates and downslope mass wasting probably in the form of turbidity currents and debris flows It is likely therefore that the gully or channel systems illustrated in Figure 2 were formed by mass flow activity under full glacial conditions The Abyssal Plain in the Bellingshausen Sea The ship traversed twice across the Bellingshausen Sea and eastern Amundsen Sea in deep water approximately parallel to the shelf edge on its way to and from Pine Island Bay Where possible the inward and outward tracks were contiguous to provide 13 overlapping swath bathymetric coverage Three main types of feature were observed during the crossing of the Bellingshausen Sea Bedrock Ridge We mapped a bedrock ridge presumably of tectonic origin trending SSW NNE running to the edges of our mosaic between about 70 31 S 102 55 W and 70 12 S 102 40 W The ridge is broken into two parts each with the same trend The gap is at about 3 400 m the approximate level of the abyssal plain The ridge extends about 800 m above the plain to 2 600 m There is an indication that the plain is higher on the east than on the west side of the ridge The single strong surface return on TOPAS records confirms that this is a bedrock feature Sediment Waves We observed a field of well developed sediment waves on the abyssal plain at a w
28. achieved in the Simrad Neptune software or if a third party package will have to be used 108 Poor quality of manuals integral help The manuals that describe the Neptune software and the installation of the EM2000 are very poor The written English is difficult to understand and is often completely ambiguous Description of the file formats is not good and procedures e g for incorporating best navigation data are not described Indeed the documentation is actually so poor that I believe that there is a strong argument that Kongsberg should be requested to update them For example after queries to Simrad head office we were supplied with two work arounds to overcome particular problems we experiecing Firstly to prevent overwriting of raw files during import replay we were advised to use command files run from the MS DOS command line as a batch file bat For example cat 0012 20030304 153455 raw all handleEmX M309 I 0012 20030304 153455 cat 0012 20030304 160531 raw all handleEmX M309 I 0012 20030304 160531 cat 0012 20030304 163608 raw all handleEmX M309 I 0012 20030304 163608 cat 0012 20030304 170645 raw all handleEmX M309 I 0012 20030304 170645 cat 0012 20030304 173722 raw all handleEmX M309 I 0012 20030304 173722 cat 0012 20030304 180758 raw all handleEmX M309 I 0012 20030304 180758 cat 0012 20030304 183835 raw all handleEmX M309 I 0012 20030304 183835 Similar
29. and location swell period for each of the Autosub Under Ice missions during JR84 In Table 11 No swath data implies a mission where swath data were collected but from the preliminary analysis it is unlikely that any of the data will be of any use see Issues with EM2000 and Processing Software Issues with EM2000 and Processing Software 1 A feature of the EM2000 and associated processing software is that it filters out what it considers to be spurious returns These returns include signals received outside of a specified valid time period see also sections by Autosub team and DGV The time period is calculated based on the sound speed and an estimate of the travel distance For example any signal that apparently appears above the sea surface is coded as a zero i e missing data at this stage it is not clear whether these data are recoverable but it seems unlikely 2 Figure 32 shows an example of the correlation plot produced by Neptune When operating correctly the EM2000 uses two methods for detecting the seabed or as in this case sea surface echo in the receiver time series Near to the centre of the swath the echo is detected using an amplitude threshold Further from the centre of the swath where the sea surface echo is buried in noise a correlation phase detection is used It is highly unlikely that amplitude detection could be successfully applied to the outermost beams As such a normal arrangement of amplitude and phase dete
30. case but our current best assessment is that this is the case This is however unsatisfactory in that while in upward looking mode it is entirely possible for a valid return to appear to be above the sea surface For example this could occur either because the input sound velocity is too great or because waves temporarily mean that the sea surface really is above the mean sea level This appeared to be a significant problem for some of the upward looking missions and remedial action was considered necessary Our remedy was to alter the setup to introduce a Depth Offset Correction DSO of 5 m and later 15 m This should mean that the EM2000 calculates the vehicle as being 5 m below its actual depth and give us 5 m of extra window the extra 5 needs to be removed in post processing and this is true of the data from mission M321 Since the introduction of the DSO appeared to work in mission M321 but not on later missions means that we re at a loss to know if this was the entire problem Furthermore the is ambiguity in the documentation regarding the sign of the DSO correction Further tests in upward looking configuration need to be completed before this issue is entirely resolved Navigation data Only the raw navigation data is passed to the EM2000 and stored alongside the raw range data in the raw files This means that the best navigation data must be merged with the raw ranges at a later time It is unclear if this can be
31. eject the flush water 3 Take Water Sample To sample 250ml two cycles of the syringe pump were required each of 125ml so the sequence take in sample move rotary valve to target port fill bag move back to port 1 was repeated twice The time taken to execute this procedure for a bag on port 25 took approximately 25 minutes Therefore a sample frequency of 30 minutes was implemented using the AquaLab in an autonomous operation mode rather than one that received commands from the Autosub mission control Because on occasion the AquaLab seemed to start up in an undefined way and failed to start its sampling script it was decided to manually start the AquaLab by command at the beginning of each mission The AquaLab would then work on its own internal timer to take subsequent samples on a 30 minute basis It should also be noted that the demanded sample of 250ml generally resulted in a sample size of between 260ml and 265ml when the AquaLab was run on deck Results Disappointingly the AquaLab failed to provide a reliable consistent and uncontaminated series of water samples On recovery of the Autosub the samples were found to be of varying volumes some greater and some less than the requested 250ml In addition it was apparent that during the sampling process many of the bags had their 50ml primed water extracted It was not possible to ascertain whether this 78 missing water was ending up as the extra volume in the sample bags Alt
32. file also with extension cnv but with a different filename This series of files were then copied to the UNIX system using samba SBE35 High precision thermometer Every time a water sample is taken using the rosette the SBE 35 recorded a temperature in EEPROM This temperature was the mean of 10 1 1 seconds recording cycles therefore 11 seconds data The thermometer has the facility to record 157 measurements but we downloaded the data approximately every 5 casts 60 measurements using the Sea Bird Electronics Inc Terminal programme Data were converted to temperature using the Sea Bird calibration routines 10 In n a In n a In n a 27345 and t slope x t offset and n is the output from the SBE 35 the other constants are listed in the appendix Salinity Samples Either six or twelve salinity samples were taken from each CTD station throughout the cruise giving a total of 411 samples with 36 duplicates The salinity samples were taken in 200 ml medicine bottles each bottle being rinsed three times before being filled to just below the neck The rim of the bottle was then wiped with tissue a plastic seal inserted and the screw cap replaced The salinity samples were then placed close to the salinometer sited in the chemistry lab and left for at least 24 hours before measurement This allowed the sample temperatures to equalise with the salinometer The samples were then analysed on
33. fixed surface such as an ice shelf at anticipated ranges of up to 200m If bottom or top surface tracking data is not available from either ADCPS then a water track velocity from the bin nearest to the Autosub is used For navigation data from the downward ADCP is used in preference to the upward ADCP As well as velocity information both ADCPs also provide ranges from each of the four beams These ranges are used by the flight control system for constant distance flight of the Autosub from either the upper or lower surface and are used to determine when the thickness of water within an ice cavity is less than preset bounds thence triggering a retreat out of the ice cavity The ranges are quantised at 196 of the range and are scaled by a factor of cosine 30 degrees to make a crude correction for the 30 degree beam angle however no correction is made for the pitch and roll of the vehicle Both ADCPs are triggered to operate on a two second cycle by a network time synchronization message sent from the mission control node to prevent acoustic interference between they and the other sonars Within each 2 second period the ADCPs transmit one bottom track and one water track ping Both the upward downward ADCPs were configured with 15 of 8 m water track bins Performance An early issue needing resolution was the function and performance of the navigation system given the high operating latitude that is known to affect the accuracy of
34. for to nose None M309 replacement of 2 2kg in water 1 1 kg added aluminium cable tray to tail 10 03 03 None Rate of ascent found to be marginal and None M312 considered to be nose heavy 13 03 03 2 2kg moved Ballast removed to increase buoyancy from nose to Network cable added to eliminate nose tail 4 M312 harness from Data Dropouts 1 1kg removed investigation from nose 4 5 1 1kg removed from tail Extension Network cable added 2 2kg in water 14 03 03 Digi Q depth Possible difficulty in diving observed sensor moved Changes made to trim slightly nose from nose to None M314 heavy and accommodate re allocation of tail 2 1kg in depth sensor and cables as data drop water out investigation continues Extension network cable removed 2 2kg removed from tail 4 5kg added to nose 19 03 03 See Remarks None M319 Nose badly damaged during recovery Extensive changes made see spreadsheet AUlantarctic2003PostRoughRecovery xls 20 03 03 None None M319 Floated in water with winched lines attached as a basic floatation test Note changes were made after a mission and so the logged dates of mission numbers and ballast change date are not necessarily the same Table 11 Record of ballast changes made during the campaign Dive Weight System The dive weight system had been developed for the AUI programme to enable the vehicle to be launched amongst floating ice without the need for run along the su
35. geology Ship s Company Officers Christopher Elliott Robert Paterson Andrew Liddell Michael Golding John Summers Charles Waddicor David Cutting Vincent Blocke Gerard Armour Steven Eadie Simon Wright Norman Thomas Kenneth Olley Crew George Stewart David Williams John McGowan Marc Blaby Derek Jenkins David Rees Lester Jolly Mark Robinson Sydney Smith Richard Turner Raymond Collins Clifford Pratley Derek Lee Kenneth Weston James Newall Ship s Doctor Alex Ramsden Master Ch Off 2 Off 3 Off Dk Off R Off C Eng 2 Eng 3 Eng 4 Eng D Eng Elect Cat O Bosun Bos Mate Sea 1 Sea 1 Sea 1 Sea 1 Sea 1 M Man M Man Ch Cook 27 Cook S Stwd Stwd Stwd Stwd Chronological outline of cruise 28 Feb 2003 JCR departed Port Stanley Falkland Islands at 09 00L We headed for 5549307 W059 35 to commence swath bathymetry and towed magnetometer section across Drake Passage The ocean logger and ADCP were also running 1 Mar 2003 We completed the Drake Passage section and commenced our transit to Gerlache Strait Underway systems were running throughout the passage 2 Mar 2003 In transit 3 Mar 2003 We arrived in Dallmann Bay early in the morning Autosub passed its buoyancy test in calm waters among the Melchior Islands The ship proceeded to Gerlache Strait for the first Autosub trial missions M307 8 We ran a swath survey in Gerlache Strait overnight 4 Mar 2003 The swath survey finished
36. in the Ashtec and gyro headings delta heading prange Forces delta heading to lie around zero Output files 84ash mrg 64 ashexec2 Complicated exec as it edits the merged data file Steps datapik2 rejects all data outside the following limits heading outside 0 and 360 pitch outside 5 and 5 roll outside 7 and 7 attf outside 0 5 and 0 5 mrms outside 0 00001 and 0 1 brms outside 0 00001 and 0 1 delta heading outside 5 and 5 pmdian removes flyers in delta heading of greater than 1 from a 5 point mean pavrge sets the data file to be on a 2 minute time basis phisto calculates the pitch limits datpik further selection of bad data outside the following limits pitch outside the limits created mrms outside the range 0 to 0 004 pavrge again sets the data file to be on a 2 minute time base pmerge merges the heading data back in from the master gyro file pcopya changes the order of the variables Output files 84ash edit 84ash ave A manual editing procedure was then performed as described in the ADCP data processing report Gyrocompass The gyrocompass is a fundamental data stream It is used by the RVS program bestnav to derive dead reckoning in the absence of GPS data as well as being used for ADCP processing ADCP report and derivation of true wind velocity ocean logger report For JR84 the gyrocompass data was read in 12 hour time periods using th
37. in the morning ready for a further Autosub trial M309 run A CTD test cast 001 was run during the Autosub mission After the vehicle was recovered the ship left Gerlache Strait via Neumayer Channel to start the transit to Peter I Island 5 Mar 2003 In transit Underway systems logging 6 Mar 2003 In transit Argos drifters were activated on the after deck 7 Mar 2003 The ship passed Peter I Island early in the morning and headed for sea ice north of Thurston Island 8 Mar 2003 We found the ice edge at first light and located a suitable site for an Autosub deployment S70 27 W102 09 The aim was to test the vehicle to full depth in open water before commencing the sub ice missions Mission M310 was run coincident with CTD cast 002 The Autosub mission was aborted part way through and the vehicle was recovered to investigate the problem The ship headed into ice to deploy the first drifter on a floe at 5709597 W101 10 Snow and ice thickness were measured on the floe The ship then returned to the ice edge and ran a swath survey overnight 9 Mar 2003 A further Autosub test mission M311 in the morning revealed a continuing problem and was aborted almost immediately We entered the ice again to deploy the two remaining sea ice drifters and sample the floes near S70 55 W102 10 The ship once again returned to the ice edge and ran a further swath survey overnight 10 Mar 2003 Autosub test mission M312 failed at depth
38. is defined by the direction normal to the 1 4 baseline i e that baseline has Y 0 Vector X R Y F Z U 1 2 2 938 4 748 0 027 1 3 1 478 4 749 0 011 1 4 13 210 0 0000 0 036 Offset 0 0 O R Max cycle 0 2 cyc smoothing N Max mag 0 08 Max angle 10 Table 8 63 The Ashtec functioned well during JR84 apart from a number of periods when no data was received see Table 9 for times and durations This was very unfortunate because of the implications for ADCP processing It also could have been easily avoided if we had maintained regular watches Day Time read 059 11 02 35 5 2 060 06 53 58 4 3 063 06 50 54 3 6 067 06 50 26 5 0 068 02 32 23 34 0 076 17 24 22 9 2 081 05 36 18 8 5 081 18 10 15 21 1 088 05 25 08 6 1 091 04 13 38 9 9 092 05 10 58 8 0 093 03 44 15 10 9 Table 9 Our complex data processing is designed with using the Ashtec to correct the gyrocompass error in mind There are were three execs involved in the processing ashexec0 1 and ashexec2 ashexecO Reads in data from the GPS3DF into pstar format Steps datapup transfers the data from RVS binary files to pstar binary files pcopya resets the raw data flag on the binary file pheadr sets up the header and data name of the file Output files 84ash raw ashexec1 Merges Ashtec data to master gyro file from gyroexecO Steps pmerg2 merges the Ashtec file with the master gyro file parith calculates the differences
39. sample bottles three times before drying the top of the bottle with a tissue and then screwing down the cap Bottles were again sealed by stretching parafilm around the neck The number of samples and their sample numbers are given in table 4 below Table 4 Samples taken for Ra analysis on cruise JR84 CTD Number Sample Station of Samples Number 001 12 84ctd001 1 12 002 12 84ctd002 1 12 008 6 84ctd008 1 3 5 7 9 11 38 CTD Data Processing In the following notes the term CC refers to the cruise number and the term NNN refers to the event number The CTD data are recorded using the Seabird data module seasave The raw data files created are CCctdNNN dat raw data file CCctdNNN con configuration file CCctdNNN bl bottle information file CCctdNNN hdr header information file 1 To process the data in the ctd unit The raw data are stored as binary files These must be converted to ASCII data files for further processing with the UNIX CTD scripts The programs used are Data Conversion module This program converts the binary file to ASCII Although it can be used to derive variables we only use it to convert the file our further processing being carried out in UNIX The output file is in the format CCctdNNN cnv Cell Thermal Mass module This program takes the output from the datcnv program and re derives the pressure and conductivity to take into account the temperature of the pr
40. ship s CTD system when a comparative CTD cast was made It was therefore replaced with an older spare serial number 2179 which gave values much closer to those of the other sensors An SBE 43 oxygen sensor Wetlabs Wetstar fluorometer and Wetlabs AC3 transmissometer were also fitted Figure 20 PRIMARY amp STARBOARD SID PORT SID Figure 20 Configuration of CTD and associated sensors for missions 307 to 319 inclusive After mission 310 the mating part of the Burton connector was found to have leaked and was replaced by an Impulse 7 way connector An appropriate connecting cable was also made up This was later modified in an attempt to rectify other network problems with the vehicle Following a difficult recovery after mission 319 the transmissometer was lost and the fluorometer was so damaged that it was not used on subsequent missions The tubing on the input to the primary temperature sensor number 2342 was also replaced with a modified version figure 21 81 STARBOARD SIDE PORT SIDE Figure 21 Configuration of CTD and associated sensors after mission 319 Software This was the first opportunity to try the recently introduced Windows version of the Sea Bird processing software Unfortunately this rejected data which had previously been accepted by the DOS version of the processing software Missions 307 to 319 were therefore processed using the DOS software The header produced by the Autosub Logger Fil
41. snagged by the damaged shell but remained intact The sub sustained serious damage No dropouts Seapam yet to be investigated Damage assessment 16 17 18 Transmissometer lost fell out as sub lifted out of water Primary CT sensor swinging under sub during recovery both C amp T damaged but look repairable SeaPam transducer hanging under sub during recovery looks OK but need to check lead Fluoremeter connecting tube broken Edgetech transmitter transducer oil filled boot ripped off ingress of water may have seriously damaged ceramic element and tuning coil rinsed with fresh water oil impregnation possible saved ingress Edgetech receiver array port torn from back plate but may be OK Mesotech forward looking sonar hit by hard object but probably OK Top panel minor damage Port panel badly damaged Stbd panel minor damage Bottom panel badly damaged Frame work badly damaged in tapered section lesser damage on parallel section apart from lower port junction with tapered section This is possibly where propeller impact occurred to be assessed Seabird connectorsn damaged by being forced backwards into domes Oxygen and C2 cables damaged SeaPam damaged internally broken ferrites 96 Mission 320 20 03 03 1 Found reasonable clearing in light ice 1 Recovered with difficulty Spiral down to 300 m 2 Mission completed successfully 2 Sub rapidly becam
42. than those further inshore on the shelf It appears that grounded glacier ice reached the shelf edge presumably at the Last Glacial Maximum 12 The main morphological features on the upper slope are gullies or channels some up to about 100 m deep Fig 2 Our preliminary use of the terms gully and channel is interchangeable this will be refined on further analysis of the features concerned The gullies are not present along the whole of the shelf edge off Pine Island Bay Instead they are found mainly along two stretches each of about 50 km width around 113 20 W 108 40 W The more westerly system of gullies is located offshore of the cross shelf trough Thwaites Trough described in the previous section At the shelf break there is a gully about every kilometre This gully or channel system forms an arborescent network The system is of stream order four that is the network has a maximum of four confluences The longest trunk channels are at least 30 km in length The second more easterly channel system has straighter channels which coalesce to form a channel system of only stream order two on the upper slope Between the channel systems the shelf edge is much less dissected by more isolated and less well developed channels On the mid slope there is some evidence of chutes and wider channels Fig 2 The largest channels occur at about 107 W in water of at least 3 500 m in depth This indicates that downslope
43. transfer of sediment takes place from the continental slope to the abyssal plain of the Amunsden Sea No sediment cores are available to provide sedimentological or chronological data on the timing of gully development or activity However modern oceanographic data suggest that mixing between dense cold water produced during sea ice formation in Pine Island Bay and the warmer waters of the Amundsen Sea will act against the production of large quantities of very dense bottom water Thus the predominant character of water flow at and beyond the shelf break is along slope and large amounts of downslope flowing bottom water are probably not responsible for channel development although analysis of CTD data collected from JR84 will provide a more detailed view of this It appears more likely that the systems were formed when glacier ice filled Pine Island Bay at the Last Glacial Maximum and grounded ice reached the shelf edge The streamlined sedimentary bedforms observed in both our work and that of Lowe and Anderson 2002 confirm that grounded ice was present on the outer shelf Fig 3 These bedforms also suggest that fast glacier flow may have been taking place along cross shelf troughs at about 113 and 108 W Thus during full glacial conditions the shelf and upper slope would have been considerably higher energy environments than the interglacial setting of today when the fronts of Pine Island and Thwaites glaciers are about 500 km distant
44. transmitting properly for these missions Beam No 60 70 80 90 100 110 120 n gene S A EP m ray 5 pL ieri a I m HE P 8 Le oa PEL Figure 32 Example Correlation Plot from Mission 321 114 Mean depth 64 Planning Normal Line Active Line Polygon Figure 33 Results of Raw Multibeam Sonar Analysis of Ice Draft for Autosub Mission 321 W105 80 W105 75 W105 70 70 96 2967046 570 97 2467046 w105 80 W105 75 W105 70 Mean depth 64 1 25000 Planning Opr clb Normal Line Active Line Polygon miaainon121 Figure 34 Results of Raw Multibeam Sonar Analysis of Ice Draft for Autosub Mission 321 for Survey Area 115 Table 11 Summary of Swath Bathymetry Data from Autosub Under Ice Missions At Deployment At Recovery Mission Date amp Longitude Latitude Swell Notes Number Time GMT of Period Deployment seconds and Recovery 321 17 15 22 3 03 105 44 3 70 58 5 10 12 Provided large number of valid data points There did not appear to be any ice over about 3metres draft which is surprising as the sea ice TO TO TO observations had identified ice thicker than this in the survey region Further analysis of this mission is required 23 30 22 3 03 105 55 83 70 49 77 Figure 33 shows the results of the raw data from mission 321 as well as the track of the vehicl
45. 03 00 00 15 03 2003 0000 2003 200300 00 25 03 2003 00 00 30 03 2003 00 00 Date and Time Figure 11 Conductivity offsets versus time for JR84 44 0 005 0 004 4 0 003 4 A t z 4 nnn 0 002 att fi _ 0 001 4 s 33 33 ts 5 tet 3 t 6 57 7 B g 01 2 t gt v 1 5 8 266 50 29 poao 150 200 2500 3000 3500 4000 2 LI v X gt ii 3 0 001 s 4 te 05 0 o le y 0 002 s v 8 le 0 003 0 004 0 005 0 006 Pres sure dbar Figure 12 Residual offsets of the corrected CTD data against pressure for the JR84 data set The SBE35 data On brief inspection offsets between the SBE35 thermometer and the primary conductivity cell showed that for 516 samples the mean offset was 0 00296 C with the SBE 35 being lower Comparison of the SBE35 with the secondary temperature cell showed that for 516 samples the mean offset was slightly lower at 0 00257 C Problems during JR84 Station 84ctd003 There was severe spiking in the primary circuit on the downcast between 180 and 255 dbar This depth range was removed from the downcast and copied in from the up cast Station 84ctd015 A Jellyfish was ingested by the CTD on the up cast This spoilt the calibration samples and also forced a thorough cleaning of the instrum
46. 03 03 Continental shelf off PIB No swath Swam into ice Little bugger Not swath data because the data do not have time messages Apparently this is due to a new version of the control software being installed The old version will be restored M321 23 03 03 Sea ice zone off Thurston Island Good data recorded on Swath M322 24 03 03 Sea ice zone off Thurston Island No data from swath possibly due to reflection being above sea level M323 25 03 03 In sea ice off Thurston Island No data from swath M324 In sea ice off Thurston Island No data from swath Note a dodgy connector may also be to blame for lack of data on Missions M322 M324 This was discovered during demobilisation 112 Autosub Under Ice Missions Chris Banks The Autosub AUV provides a potentially valuable platform for measuring the draft of Antarctic sea ice Autosub was fitted with a swath multibeam bathymetry system Simrad EM2000 that could be configured either in a downward looking mode or as for the purposes of this study in an upward looking configuration In addition the upward ADCP system can be used to calculate an independent measure of ice draft In addition to measurements from the vehicle data were collected in the study regions in the form of sea ice observations see Sea Ice Observations report and video images whilst the ship was en route to the deployment sites Table 11 shows the basic information start and stop time
47. 1796 199356 55220 41 o4 001 10 18 277 0012 20030325 174255 RAW 570 29 630 W98 30 051 570 30 930 W98 30 057 1802 200022 7293 45 64 002 10 54 _ 3 66 0012 20030325 194521 RAW 570 32 079 W98 30 544 570 32 497 W98 30 792 1795 199245 6928 45 81 1 122 348 0012 20030325 181331 RAW 570 30 931 W98 30 057 570 32 130 W98 30 067 1800 199800 7612 49 93 0 02 10 1 a8 0012 20030326 002052 RAW 570 28 064 W98 30 701 570 26 572 W98 30 362 1794 199134 102468 160 56 71 07 91 02 51 46 0012 20030326 005129 RAW 570 26 572 W98 30 362 570 25 021 W98 30 001 1809 200799 104618 153 46 56 15 91 05 52 10 0012_20030325_201558_RAW _ 570 32 497 98 30 792 ____ 570 32 076 W98 31 032 1794 199134 7728 44 57 11 3 88 10 Number of times pulse read 11 Number of times pulse read multiplied by number of beams used 120 0012 20030326 022319 RAW _ 570 25043 wos so o31 7025046 1800 200020 104225 15599 912 8211 121 CTD Water Sampler and ADCP Dan Hayes and Ziggy Pozzi Walker It is the purpose of this section to discuss the quality of the CTD water sampler and ADCP data collected by the Autosub during cruise JR84 The Autosub carries a pair of pumped SeaBird Electronics Conductivity Temperature cells The SBE 911 system also includes a pressure sensor oxygen probe fluorometer and transmissometer Data are collected at 24 Hz The water sampler can collect up to 49 water samples tak
48. 2 9 2 95 3 Conductivity S m Comparison of Autosub and ship s CTD conditivities Temperature Autosub m316 and JR84ctd011 300 400 500 600 700 800 900 1000 1100 1200 H Autosub T1 Autosub T2 JCR T1 ser 2 1 0 5 0 0 5 1 Temperature degC Comparison of Autosub and ship s CTD temperatures 123 Oxygen Autosub m316 and JR84ctd011 Pressure dbar 3 5 4 4 5 5 5 5 6 6 5 7 7 5 8 Oxygen ml l Figure 37 Comparison of Autosub and ship s CTD dissolved oxygen values Temperature vs Salinity Autosub m316 and JR84ctd01 1 Autosub TS1 Autosub TS2 JCR TS1 JCR TS2 o Temperature 1 1 1 L 1 1 1 1 i 1 i 34 25 343 3435 344 3445 345 3455 346 3465 347 Salinity psu Figure 38 Comparison of Autosub and ship s CTD temperature salinity values 124 Another potential improvement is the minimization of frictional heating of the water sampled for temperature The intake ports protrude from the hull near the nose so they are already away from the boundary layer on the sub s surface The piping to the sensor should be entirely vinyl or plastic and currently the intake pipe is metal This effect should be very small The oxygen probe is unfortunately in disagreement with both shipboard probes which have problems of their own It appears that the probe shows reasonable changes throughout the water column however the absolute values seem
49. 22 113 46 31 71 42 73 594 41 999 84ctd019 17 03 2003 076 06 01 113 7878 71 7075 113 47 27 71 42 45 594 571 6 64 84ctd020 17 03 2003 076 07 36 113 5234 71 6871 113 31 4 71 41 23 631 610 5 62 84ctd021 17 03 2003 076 09 07 113 3254 71 6319 113 19 52 71 37 92 636 617 2 81 84ctd022 17 03 2003 076 10 34 113 1269 71 5787 113 7 61 71 34 72 620 618 7 28 84ctd023 17 03 2003 076 12 00 112 8989 71 5509 112 53 93 71 33 05 541 518 7 37 84ctd024 17 03 2003 076 13 32 112 6505 71 5187 112 39 03 71 31 12 442 426 3 13 35 84ctd025 17 03 2003 076 17 33 113 9987 71 291 113 59 92 71 17 46 2028 2044 8 06 84ctd026 18 03 2003 077 03 18 113 7842 71 3764 113 47 05 71 22 58 1556 1485 7 13 84ctd027 18 03 2003 077 05 20 113 6346 71 4221 113 38 08 71 25 33 1090 1011 2 47 84ctd028 18 03 2003 077 07 00 113 5368 71 4573 113 32 21 71 27 44 651 631 8 35 84ctd029 18 03 2003 077 08 34 113 3675 71 5257 113 22 05 71 31 54 637 626 6 4 84ctd030 19 03 2003 078 14 32 113 6786 71 2478 113 40 71 71 14 87 0 1997 7 96 84ctd031 19 03 2003 078 16 53 113 4769 71 3399 113 28 62 71 20 39 1543 1535 8 74 84ctd032 19 03 2003 078 18 35 113 351 71 3933 113 21 06 71 23 6 1030 946 8 11 84ctd033 19 03 2003 078 20 36 113 2656 71 4102 113 15 94 71 24 61 776 782 8 28 84ctd034 19 03 2003 078 22 02 113 2112 71 4275 113 12 67 71 25 65 642 632 7 91 84ct
50. 23 EDF 11 30 072 71 02 63 S 108 28 82 W 2077 JR84 20 asvp 21 T5 0024 EDF 19 34 073 7049 43 S 106 54 14 W 2840 JR84 21 asvp 22a T5 0025 EDF Failed wire snap 22b T5 0026 EDF 00 32 075 71 00 64S 111 44 38 W 2709 JR84 22 asvp 23 T7 0027 EDF 18 00 076 71 35 84S 113 28 12 W 621 JR84 23 asvp 24 T7 0028 EDF Failed wire snag 25 T7 0029 EDF 07 00 076 71 42 72S 113 39 27 W 590 JR84 24 asvp 26 T7 0030 EDF 08 50 077 71 31 56 S 113 21 75 W 624 JR84 25 asvp 27 T7 0031 EDF 00 37 078 71 30 04S 112 49 04 W 422 JR84 26 asvp 28 T7 0032 EDF 06 25 079 7128 24 8 114 48 74 W 963 JR84 27 asvp 39 T5 0033 EDF 07 55 080 71 20 18S 114 05 77 W 1900 JR84 28 asvp 30 T5 0034 EDF 00 22 082 Failed 31 T5 0035 EDF 00 35 082 70 42 63S 105 50 86 W 3111 JR84 29 asvp 32 T5 0036 EDF 10 58 084 70 11 99S 98 21 78 W 4046 JR84 30 asvp 33 T5 0037 EDF Failed wire snap 34 T5 0038 EDF 17 30 086 68 33 88S 84551 22W 3680 JR84 31 35 T7 0039 EDF 16 15 087 67 40 77S 73 07 19W 485 JR84 32 asvp 25 SIMRAD EA500 Bathymetric Echo Sounder Mark Brandon Open University The RRS James Clark Ross is equipped a Simrad EA500 echo sounder with the transducer mounted on the hull just to starboard with the primary visual display and controls located on the bridge The system was run virtually conti
51. 59 26 79 6 37 92 0015 20030323 201530 570 37 768 1 102 38 113 S70 36 074 102 37 549 1796 199356 77776 140 26 62 2 79 99 39 01 0015 20030323 204606 raw S70736 074 W102 37 549 S70 34 370 _W102 37 118 1806 200466 77310 143 5 59 2 80 13 38 57 0015 20030323 211643 570234 370 W102 37 118 S70 32 729 102236 893 1813 201243 76012 140 02 56 82 80 03 37 77 0015 20030323 214720 570232 729 102 36 893 S70 31 124 102236 316 1802 200022 75002 139 78 58 46 80 72 37 50 0015 20030323 221757 raw S70731 123 W102 36 316 S70 29 610 102 35 522 1796 199356 78738 139 73 47 8 79 58 39 50 0015_20030323_224833_ 570 29 610 _ 102 35 522 570 27 982 W102 35 139 1798 199578 73255 142 86 61 52 80 73 36 70 0015 20030323 231910 570 27 982 W102 35 139 570 26 513 W102 34 355 1796 199356 71976 144 49 53 64 79 53 36 10 0015 20030323 234947 570 26 513 102 34 355 570 25 050 _ W102 33 482 1796 199356 79546 140 78 59 83 79 97 39 90 0015 20030324 002024 raw S70 25 050 _ 102 33 482 570 23 572 102 32 577 1788 198468 79588 139 83 59 92 80 28 40 10 0015 20030324 005100 570 23 572 W102 32 577 S70 22 088 1 102231 687 1801 199911 78332 140 37 58 73 79 42 39 18 6 Number of times pulse read 7 Number of times pulse read multiplied by number of beams used 118 Summary of Performance of Mission 323 Line Start Posn Start Posn End Posn End Pos
52. 7 12 84ctd033 12 84ctd034 2 6 8 12 84ctd035 84ctd036 10 12 84ctd037 8 9 10 11 12 84ctd038 6 7 10 84ctd039 12 84ctd040 2 12 84ctd041 9 12 84ctd042 8 12 84ctd043 11 84 044 7 9 11 42 A real problem was during the analysis of the salinity samples for stations 84ctd033 to 84ctd037 shaded in the table above Here the salinometer lab appears to have heated up during the analysis and unfortunately the person making the measurements neglected to run a standard through at the end of the analysis The safest thing to do was to ignore these samples completely Figure 10 shows the primary conductivity offset against station number figure 8 with stations 33 to 37 excluded Figure 10 suggested that a constant correction was perhaps not the best way forward for calibrating the CTDs When plotted against time Figure 11 a similar pattern emerges and so it was felt that a time dependant offset would provide a better calibration The calibration values applied to each cast are shown in the table 6 below NB these values have only been applied to the PRIMARY CONDUCTIVITY CELL Table 6 CTD calibration offsets applied to the Primary Conductivity cell for JR84 CTD Number Correction CTD Number Correction 84 001 0 00508 84 023 0 00686 84ctd002 0 00563 84ctd024 0 00687 84ctd003 0 00592 84ctd025 0 00690 84ctd004 0 00636 84ctd026 0 00695 84ctd005 0 00641 84ctd027 0 00696 84ctd006 0 00642 84ctd028 0 00697 84ctd007 0 00643 84ctd029 0 0069
53. 8 84 0008 0 00644 84ctd030 0 00716 84ctd009 0 00659 84ctd031 0 00717 84ctd010 0 00661 84ctd032 0 00718 84ctd011 0 00669 84ctd033 0 00719 84ctd012 0 00670 84ctd034 0 00720 84ctd013 0 00671 84ctd035 0 00721 84ctd014 0 00672 84ctd036 0 00727 84ctd015 0 00673 84ctd037 0 00728 84ctd016 0 00681 84ctd038 0 00729 84ctd017 0 00682 84ctd039 0 00730 84ctd018 0 00683 84ctd040 0 00731 84ctd019 0 00683 84ctd041 0 00742 84ctd020 0 00684 84ctd042 0 00775 84ctd021 0 00685 84ctd043 0 00788 84ctd022 0 00686 84 044 0 00801 After the conductivity offset was applied as per the description above the samples were merged with the corrected CTD data and new corrected sample files derived Figure 12 shows the residual offset of 251 samples against pressure for cruise JR84 There is no apparent pressure effect with the residuals shown in figure 12 and the mean offset for the corrected data against the CTD data is 0 0000 with a standard deviation of 0 0013 43 0 00900 0 00800 0 00700 0 00600 0 00500 o e e CTD offset from sample 0 00300 0 00200 0 00100 0 00000 JR84 CTD offsets Vs Station number Station number Figure 10 Conductivity offsets versus station number for JR84 0 00900 0 00800 0 00700 0 00600 0 00500 Offs et 0 00 400 0 00300 0 00200 0 00 100 0 00000 JR84 CTD offsets vs Time 28 02 2003 00 00 05 03 2003 00 00 10 03 20
54. Autosub taking longer than planned Floes were selected based on their suitability safety for working on As such the sampled floes were larger in area and were likely to be thicker i e more stable than most within the region On the floe on 8 March Floe Station 1 FS1 the OU auger was used this proved ineffective at drilling through the ice and a large amount of time was spent digging the auger out when it became stuck The ice could only be measured as greater than 1m as it would not have been prudent to use the extension rods to drill further Snow depth measurements were made and produced a range of depth between 6cm and 95cm with a mean and standard deviation of 47 4 21 1 n 103 The floe was occupied from 00007 i e actually on the 9 March until 0050Z Longitude was 101 09 76 and latitude was 70 58 84 FS2 on the morning of the 9 March used the JCR s ice anchor drill this proved more efficient compared to the previous day s drill The depth of the sea ice not including snow depth was 1 91m Again snow depth measurements were made across the floe Range of snow depth values was from 5cm to 88cm with a mean and standard Worby 1999 Observing Antarctic sea ice A practical guide for conducting sea ice observations from vessels operating in the Antarctic pack ice A CD ROM produced for the Antarctic Sea Ice Processes and Climate ASPeCt program of the Scientific Committee for Antarctic Research SCAR
55. BE 43 2 Serial number 0242 Calibrated on 27 Aug 02 Soc 4 5920e 001 Boc 0 0000 Offset 0 4597 Tcor 004 0001 Pcor 1 35e 004 Tau 25 50 220 11 Voltage channel 5 Free 12 Voltage channel 6 Fluorometer Chelsea Aqua 3 Serial number 088216 Calibrated on 11 june 01 VB 0 260700 V1 2 035000 Vacetone 0 326300 Scale factor 1 000000 Slope 1 000000 Offset 0 000000 13 Voltage channel 7 Free Oceanlogger Underway Measurements Mark Brandon Open University Throughout JR84 underway measurements were made with the ship s oceanlogger The oceanlogger system is comprised of a thermosalinograph and fluorometer connected to the ship s non toxic pumped seawater supply plus meteorological sensors measuring duplicate air pressure duplicate air temperature duplicate humidity duplicate total incident radiation TIR and duplicate photosynthetically available radiation PAR There were 18 sensors logged in total within the oceanlogger system To complete the meteorological data set I merged in the windspeed and direction from the anemometer Data are time stamped using the ship s master clock Calibration details Up to date calibration certificates for all sensors was provided by the Pat Cooper ETS Data Processing Oceanlogger data were processed in 12 hour segments throughout the course of JR84 Three Unix scripts calling PSTAR software routines were used for this processing 84 1
56. DCP data The figure also shows that the data collected with the upward looking ADCP at 90 m depth are apparently valid for the bins 9 11 On further examination it seems that this data may be contaminated by surface echos In any case the range to the tracked surface appears to be very good when within 100 m When combined with the depth of the vehicle the range can be used to calculate ice draft a short example of which is also shown Note the different horizontal scale Another problem with the downward ADCP is that the range data sometimes indicate an approaching obstacle and cause the collision avoidance algorithm to be activated many times during the run every 150 seconds or so in m324 This can cause extreme pitch angles 10 to 25 degrees and depth changes up to 9 m that must be carefully corrected for in the ice draft calculation VEdn VNdn Time s x10 Figure 39 Autosub ADCP data for M323 127 222222224 uem 3 6 3 8 4 4 2 44 46 Distance Travelled km Ice Draft m A 4 io A Figure 40 Autosub position and derived ice draft during M323 128 Cruise Track Plots Toby Benham Scott Polar Research Institute JR84 Cruise Track Maps Ke Legend Legend of Year Symbology 59 CTD cast station 60 CTD cast number jr84 1 Hour of day 24hr CT r Start of Day Marker 58
57. Difference is 10 031000 seconds Time at server Asublog1 was 04 03 2003 20 54 57 when time at host Autosub9 was 04 03 2003 20 54 53 Local time is 20 54 43 619 Difference is 4 000000 seconds 50 Fsau is 8 584 s faster than local Autosub9 is 10 031 s faster that local and Asublog1 is 4 s slower than Autosub9 Therefore Asublog1 is 10 031 4 faster than local FSau is 8 584 10 031 4 faster than Autosub9 ie Nav 5 44 sec Find time corrections for each file I used excel to calculate a single time corrections for each jsf file by interpolating between the mission start and end corrections I did this in Excel Correction Mission start 13 34 58 0 075 Mission end 20 54 50 5 44 Opened file data0001177 jsf16 40 47 0 2 34138 16 50 05 0 2 45481 Opened file data0001178 jsf16 50 05 1 2 45481 17 00 51 1 2 58613 Opened file data0001179 jsf 17 00 53 2 2 58654 17 01 01 2 2 58817 Opened file data0001180 jsf 17 30 05 3 2 94269 17 40 08 3 3 06527 Opened file data0001181 jsf 17 40 09 4 3 06547 17 50 03 4 3 18622 Rounding to one half second precision implies that a single correction of 2 5 should be used for the first segment and 3 sec for the second set Although depth data is available only at one second intervals the interpolation of the depth as recorded in the bnv file allows a correction that is less than one second 102 Apply Autosub time depth corrections Now I used Matlab routine depthinterp m to apply cl
58. Differential mode DGPS During JR84 the DGPS reference station at Stanley was used Instrument Type Code Use Trimble 4000 GPS receiver gps Primary positional information Ashtec GG24 GLONASS GPS receiver glo Positional information Ashtec ADU 2 GPS receiver ash Attitude information Gyrocompass Sperry Mk 37 model D gyr Heading information Electromagnetic Log Chernikeeff log Aquaprobe Mk V eml Velocity information Table 7 The collection and use of all of the navigation data are linked of the instruments are currently logged to the SCS system and then transferred to the old RVS Level C system where they are currently read During cruise JR84 the data for all five instruments and the standard editing procedures were done in one Unix script called jr84 go This script requires the Julian day and am or pm selection as input and then executes a further 8 C shell scripts to read in 12 hours of data and edit where necessary all five streams This report briefly describes each instrument and explains the processing as was performed on cruise JR84 The instruments Trimble 4000 The Trimble 4000 receiver in differential mode was the primary source of positional information for the scientific work on JR84 The data were logged at 1 second intervals and read into pstar files in 12 hour periods from the SCS derived Level C stream using the Unix script gpsexec0 Individual steps in this exec are as follows 62 gpsexecO
59. JR84 Cruise Report Autosub Under Ice Cruise to the Amundsen Sea RRS James Clark Ross 28 February to 4 April 2003 Report complied by Adrian Jenkins from the contributions of the scientific party Chris Banks Toby Benham Mark Brandon Jon Copley Julian Dowdeswell Jeff Evans Sarah Hardy Dan Hayes Adrian Jenkins Steve McPhail Nick Millard Colm O Cofaigh Miles Pebody James Perrett Ziggy Pozzi Walker James Riggs Pete Stevenson David Vaughan and Andy Webb Contents Introduction Summary Scientific Party Ship s Personnel Chronological outline of cruise Sea ice conditions in the Amundsen Sea during JR84 Outreach activities Marine Geophysical Investigations EM120 TOPAS EPC chart recorder XBT s Simrad EA500 Bathymetric Echo Sounder Sea Ice Observations Sea Ice Drifters Physical Oceanography JR84 CTD operations SBE43 dissolved oxygen sensors Appendix Calibration data Oceanlogger Underway Measurements ADCP Measurements Summary The configuration of the ADCP Standard method of processing Identification of CTD on station ADCP data Navigation Data The instruments Daily navigation processing Microbiological Sampling Autosub Operations Trim and ballast Edgetech FS AU sub bottom profiler Simrad EM2000 multibeam swath system AquaLAB Autosub SBE 9 CTD ADCPs and navigation Mission descriptions Connector problems Damage sustained during recovery after mission 319 Summary of problems encountered during c
60. Phail Miles Pebody James Perrett James Riggs Pete Stevenson Andy Webb Southampton Oceanography Centre Mobilisation Autosub its launch and recover gantry ancillary equipment and battery boxes were loaded into 3 x 20 foot containers for shipping from Southampton to the Falkland Islands on the 10 December 2002 and were awaiting the ships arrival in Stanley on 24 February 2003 Mobilisation began on the Afternoon of 25 February and continued until the ship sailed on the 28 February Two of the transportation containers had been modified to double as garage space and workshop for Autosub and were located on the aft deck just forward of the launch and recovery gantry on the port quarter The third open top container used to ship the gantry and was unloaded and left ashore The fish containing the tracking and telemetry acoustics was loaded on to the PES winch on the starboard side just forward of the bridge Container workshop garage Experience gained from working on the vehicle on the open deck in the Weddell Sea 2 years previous highlighted the need for a warm storage and working environment for Autosub and the engineering team To this end two of the shipping containers were modified to be used as a workshop and garage The workshop container was a refrigeration container and as such well insulated fitted with a personnel door and window A lifting beam with hoist ran its length to assist dismantling Autosub and moving heavy items
61. S degrees cm s cm s cm s cm s m degrees C degrees S degrees cm s cm s cm s cm s m degrees C degrees cm s cm s 58 Identifying CTD on station ADCP data A CTD station was selected from those shown in Figure 15 below and the corresponding Julian date and time am or pm were identified from the CTD log CTD station positions 1 12 11 3 3 36 p 12 4 32 gs 5 37 27 70 i v 38 p i 9 39 35 29 24 5 40 i 23 22 A 21 20 48 _ 72 i 176 w 115 W 1149w 1139w 1122W Longitude Figure 15 Plot of CTD stations From the corresponding abs file the ve and vn variables ship velocity averaged over 2 minute periods in the east and north direction respectively were plotted From this plot approximate start and stop times of the period when the ship was stationary during the CTD deployment were noted Using the same abs file every 64 data cycle i e start of every 2 minute time averaged ensemble was listed using m ist and displaying the variables time JDAY bindepth absve absvn ve and vn From this list those data cycles closest to the times noted previously from the ship s velocity plot were identified The data cycle closest to the start of the stationary period with ve and vn both nearing 0 cm s was noted For the end of the stationary period the data cycle listed that clearly showed the ship to be moving off stati
62. S Long W Water JR84 asvp file depth m 1 T5 0002 EDF 15 13 060 55 24 65 S 60 17 84 W 4000 JR84 1 asvp 2 T5 0004 EDF 22 16 060 56 29 765 61 10 20 W 4370 JR84 2 asvp 3 T5 0006 EDF 07 00 061 57 52 14 S 62 28 73 W 3541 JR84 3 asvp 4 T5 0007 EDF 16 00 061 59 53 90 S 62 38 52 W 4040 JR84 4 asvp 5 T5 0008 EDF 22 06 061 61 19 54 S 62 45 66 W 3512 JR84 5 asvp 6 T7 0009 EDF 07 10 062 63 15 11 S 62 58 03 W 703 JR84 6 asvp 7 T7 0010 EDF 16 10 062 64 35 19 S 62 34 09 W 683 JR84 7 asvp 8 T7 0011 EDF 03 00 063 64 13 33 S 61 49 86 W 868 JR84 8 asvp 9 T7 0012 EDF 17 04 064 65 52 48 S 70 15 14 W 361 JR84 9 asvp 10 T5 0013 EDF 21 00 064 66 10 13 S 72 04 44 W 2926 JR84 10 asvp 11 T5 0014 EDF 08 20 065 66 55 26S 76 45 36 W 3737 JR84 11 asvp 12 T5 0015 EDF 13 15 065 67 18 31S 79 05 31 W 3917 JR84 12 asvp 13 T5 0016 EDF 19 11 065 6743 19 S 81 53 45 4008 JR84 13 asvp 14 T5 0017 EDF 09 40 066 68 49 32S 89 11 74W 3683 JR84 14 asvp 15 T5 0018 EDF 06 10 067 70 14 79S 9940 07 W 3970 JR84 15 asvp 16 T5 0019 EDF 04 35 070 70 15 58S 101 54 12 W 3710 JR84 16 asvp 17 T7 0020 EDF 20 03 070 71 14 83S 109 15 80 503 JR84 17 asvp 18 T7 0021 EDF 04 10 071 7204 01 S 111 03 39 W 563 JR84 18 asvp 19 T7 0022 EDF 19 00 071 7246 28 S 109 25 11 W 470 JR84 19 asvp 20 T5 00
63. Swath data no good Launch outside ice spiral EM beacon but not Looked like it passed 2 On 27 found transmitter lead on EM200 transmitter leaked to holding depth 100m 4 5 km abeam later found to be East 3 Strong current responsible for track error 0 3 knts mostly east run for 0 5 hr at 10 m into ice for 3 miles grid survey out to recovery point mes 6 Chased up to recovery 1 WP no sign on TPII but established circling and appeared to be East Timed out and sub headed for recovery 2 Ship set of N to try to catch up sub EM beacon showed aborted mission time out range 4 km Gonio signal located sub 8 km east of track 97 Autosub Data Analysis David Vaughan Dan Hayes Chris Banks Ziggy Pozzi Walker Toby Benham James Perrett British Antarcitc Survey Open University Scott Polar Research Institute Southampton Oceanography Centre Edgetech FS AU sub bottom profiler David Vaughan With assitance from Toby Benham James Perrett and Chris Banks The Edgetech sub bottom profiler was configured to run from Autosub in a downward looking orientation Prior to the Autosub under ice cruise it produced data during the trials in August 2002 which confirmed it s correct installation It is a valuable tool that should allow discrimination of seabed types and image sub bottom structure in sedimentary sequences During the Autosub under ice cruise it ran from M308 to M319 On the last mission
64. Underway systems logging Good speed through calm seas 4 Apr 2003 With continuing fair weather the JCR arrived in Port Stanley at 14 00L Sea ice conditions in the Amundsen Sea during JR84 Adrian Jenkins British Antarctic Survey The summer of 2003 was unusual in the Amundsen Sea in that the whole of Pine Island Bay and the continental shelf to the west was almost completely clear of sea ice This situation persisted up to the end of February when JR84 commenced Indeed at the time of our departure from the Falklands Pine Island Glacier appeared to have the most accessible ice front of any ice shelf in the south east Pacific sector of Antarctica Figure la It seemed that the main problem might be finding suitable multi year floes for the sea ice programme These autosub missions had to be run prior to those in Pine Island Bay Their shorter duration and the possibility that the vehicle might be retrievable by ship if anything failed made them an ideal lower risk test environment for many of the new Autosub features The area to the north of Thurston Island was selected for the sea ice work On arrival in this region the open conditions further south and west in Pine Island Bay appeared to be stable although a tongue of multi year ice was beginning to move west threatening to close off access to the Bay Figure 1b By the time JCR attempted to access Pine Island Bay on 12 13 March the situation was changing rapidly The tongue of ice
65. ailable as part of the package This may be a route for the future but the route described below is based on using bespoke routines in Matlab which do not provide a turn key solution and require operator intervention 98 Processing of M309 Profiler data Jstar quality assurance The Edgetech data are downloaded from Autosub in Edgetech s proprietary format jsf Each file contains around ten minutes of data and this is a satisfactory arrangement for the present although for longer missions fewer files might be easier to deal with quickly This format of the files is said to be a variant on Seg y but it does not appear to be readable by any of the standard packages that read Seg y format Thus any similarities to Seg y are not directly helpful and other arrangements to read the data are required Furthermore those data formats are not sufficiently well described in the manual and we have spent considerable time and effort in reading the data both in Matlab and The basic software package supplied with the Edgetech is sufficient however to do basic quality assurance and control of the data even though the Jstar manual is inadequate in explaining even the basic features of the software For example it should be noted that there is the facility to capture the jsf data to jpeg images allowing printout which is not described in the manual see example below It is likely that at the end of each mission Jstar sh
66. al sampling Jon Copley Southampton Oceanography Centre Water samples were collected and preserved by Copley for microbial analysis with researchers associated with the NERC Marine and Freshwater Microbial Biodiversity programme Samples for this purpose were taken using sterile 50 ml containers from Niskin bottles filled during several CTD casts For analysis of prokaryotes by flow cytometry and in situ hybridisation Zubkov SOC 2 x 12 ml subsamples were transferred into 15 ml sterile tubes using clean pipette tips and each fixed with 600 pl of 0 2 um filtered 20 paraformaldehyde For investigation of viruses by electron microscopy and molecular techniques Wilson MBA a further 1 ml subsample was transferred into a sterile 1 8 ml cryovial and fixed with 10 ul of 50 glutaraldehyde All samples were inverted several times and left to fix in the cold room before being frozen at 80 C The frozen samples will be transported to the UK in the 80 C freezer aboard the ship As Autosub did not sample water beneath an ice shelf no live microeukaryote samples were collected for Finlay at CEH The samples were collected from a range of depths during 5 CTD casts across the shelf break at the entrance to Pine Island Bay Two of these casts were to 2000 metres and three were to 500 metres including one cast below sea ice A total of 84 samples were collected and preserved for analysis of prokaryotes and viruses as shown in the summary table of mi
67. all gyro based navigation systems No problems were found with the INS up to the maximum latitude we achieved of 72 degrees south although the absence of bottom track data precluded any determination of navigation accuracy as the actual accuracy is determined solely by the magnitude of the currents which for some missions were substantial up to 0 2 m s The only adverse effect was that the INS alignment time following first GPS fix of the INS increased to about 10 minutes normally about 5 minutes To speed up Autosub deployments a GPS antenna was mounted on the roof of the Autosub container and plugged into the Autosub navigation system so that INS alignment could be completed before the Autosub was taken out of the container During the early trails in Gerlach straights we were able to check the bottom track performance of the ADCP Bottom track navigation was achieved reliably at altitudes of up to 350 m off the sea floor occasionally up to 400m which was set as the maximum range Bottom tracking navigation was probably adversely affected by the 83 relatively steep descent and ascent angles 15 degrees and so these figures can be considered to be minima Problems Throughout the cruise the downward looking ADCP gave an unacceptably high level of missed pings Typically only 75 of the expected data were recorded Unsuccessful attempts were made to track down the cause of this problem which we think lies in either the ADCP hardwa
68. ater depth of over 3 000 m They are located at about 69 S between 92 and 98 W about a 200 km length beginning about 70 km south west of Peter I Island The crests of these sea floor sediment waves have a wavelength of a few 100 m and amplitudes of up to about 5 The waves appear to be between 1 and 2 km in length However the edges of our swath coverage truncate many of them The wave crests are orientated between WMW ESE and NW SE TOPAS records have penetration of several tens of metres though the acoustically stratified sediments making up the waves The waves are presumably related to currents running along the sea floor close to the base of the continental slope Charcot Canyon We traversed Charcot Canyon twice on courses that provided sections across it in water depths of about 3 800 to 4 100 m at about 67 30 S 77 W Three types of acoustic facies are seen on TOPAS records across the canyon First there are acoustically stratified sediments with penetration of the signal often exceeding 30 m beneath the sea floor This acoustic facies is interpreted as turbidites Secondly there is a series of isolated lens shaped semi transparent units bedded within the acoustically stratified facies The lenses are up to about 5 m thick and several hundred metres wide These lenses are interpreted as debris flows Thirdly there are some small channel like features at the surface of the acoustically stratified facies These are probably submari
69. ative speed of sound 26 Sea Ice Observations Chris Banks and Mark Brandon Open University Sea ice observations were made from the bridge as progress and personnel permitted using the standardised ASPECT approach Information collected included type s of ice floe size estimate of thickness snow cover type and depth topography longitude and latitude time and basic meteorological data The observations will be validated by the use of digital photographs and video images back in the UK which will allow more accurate descriptions of coverage In addition thickness can be estimated from comparison with parts of the ship visible in the photographs There is a possibility that some of the earlier images have been corrupted this is yet to be confirmed Sea ice observations were carried out on the 8 9 10 12 13 14 15 20 22 23 24 and 25 March 2003 data were recorded on paper and then entered into the ASPECT software These data should be of particular interest because the process of sea ice formation in relatively calm water has been observed Measurements on Ice Floes Three locations were used to measure ice thickness The first site was on the afternoon of 8 March There were two sites on the 9 March one in the morning and a second in the afternoon The measurements were designed to act as ground truth for the Autosub missions Unfortunately there were no under ice missions within the next few days due to testing of
70. b is a further development of the AquaMonitor The instrument consists of a 200ml syringe type pump and a rotary valve that selects one of fifty ports Port number 1 is used to acquire water from the outside of the Autosub and the remaining 49 ports are fitted with sample bags The requirement for this cruise was to take 49 samples of 250ml each Previous problems with this instrument and with the earlier AquaMonitor lead to extensive discussion with WS Envirotech prior to the start of the cruise The programming and use of the device has proved to be complex and in the end ineffective The following describes the strategy recommended by WS Envirotech e Bags were to be filled in the following sequence Ports 25 down to 2 with anticlockwise rotary valve movement and ports 26 up to 50 with clockwise rotary valve movements This sequence was considered necessary in order to avoid cross contamination of samples held in bags as the rotary valve passes their respective port e Prime all bags with 50ml water before attaching them to the AquaLab This is to prevent undue stress on certain parts of the bags and seals at depth e Each sample process was then to proceed as follows 1 Extract the primed water Move rotary valve to desired port extract the primed water Move rotary valve to port one and eject the prime 2 Flush the sample bag with 190ml Intake 190ml water move to the target port fill and empty the bag return to port 1 and
71. ce length 400 ms file size 10 MB under processing Swell OFF Dereverb OFF Stacking OFF Shallow water settings lt 1000 m water depth Burst source period 1 3 level 100 secondary frequency 2800 Hz SSU triggering ping interval set to 0 Gain 15 25 dB depending on seabed type and conditions Processing filter ON AVC ON Scale 2000 Deep water settings gt 1000 m water depth Chirp source 15 ms pulse length 1 1 5 kHz level 85 bandpass filter settings 1400 1600 4900 5 100 Hz SSU triggering ping interval set 0 Gain 20 30 dB depending on sea bed type and conditions Processing filter ON Deconvolution ON 1 ppm TVG ON manual start about 200 300 ms above seabed slope generally ranged from 50 90 dB s AVC ON Scale 3000 As with previous cruises it was found that TOPAS produced poor returns on steep slopes and undoubtedly works best on a fairly flat sea bed where impressive records can be achieved in both Burst and Chirp modes The 400 ms trace length meant that in steep irregular topography frequent delay changes were required Post processing of selected TOPAS files was done during the course of the cruise In previous cruises and cruise reports it had been noted that it would be advantageous to have EM120 and TOPAS post processing software available on different workstations However this problem has now been alleviated following the production and installation of the new 2002 PC based TOPAS post pr
72. configured through the Direct Command menu of the DAS software using the command 0004 This sets the instrument to one bottom track ping for every four water track pings The ADCP does not log to the SCS system unlike all other underway scientific instruments on the RRS James Clark Ross but instead the 2 minute ensembles of data are fed directly into the ship s Level C system In the event of a problem with the ship s Level C system the data has to be recovered from the PC files but no such problems were encountered during JR84 51 Standard method of processing The steps involved in processing the data are detailed below and summarised in the flowchart in Figure 14 The data were read into pstar files of 12 hour periods from the Level C system and processed using the pstar processing software The programs involved also require data from several navigation streams described in the navigation data report Step 1 Reading data The data were read in and saved in 12 hour periods 00 00 to 11 59 and 12 00 to 23 59 using the Unix script 84adpexec0 This processing produces two output files one containing the water track data and one containing the bottom track data When the ADCP was set to record only water track information the bottom track file contains only engineering data and zero s for the bottom velocity Output files 84adp 3 digit Julian day plus a or p for am or pm 84bot Step 2 Water velocity te
73. crobial samples 68 Table 10 Summary of samples collected and preserved for microbial analysis Sample 25 1P 25 1V 25 2P 25 2V 25 3P 25 3V 25 4P 25 4V 25 5P 25 5V 25 6P 25 6V 25 7P 25 7V 25 8P 25 8V 25 9P 25 9V 25 10P 25 10V 25 11P 25 11V 25 12P 25 12V 30 1P 30 1V 30 2P 30 2V CTD Depth 2044 2 x12 ml prokaryotes ml viruses 1733 2 x12 ml prokaryotes ml viruses 1424 2 x12 ml prokaryotes ml viruses 1120 2 x12 ml prokaryotes ml viruses 2 x12 ml prokaryotes ml viruses 2 x12 ml prokaryotes ml viruses 2 x12 ml prokaryotes ml viruses 2 x12 ml prokaryotes ml viruses 2 x12 ml prokaryotes 1 ml viruses 2 x12 ml prokaryotes ml viruses 2 x12 ml prokaryotes ml viruses 2 x12 ml prokaryotes ml viruses 1996 2 x12 ml prokaryotes 1 ml viruses 1833 2 x12 ml prokaryotes Sample CTD Depth Description Sample CTD Depth Description m m 30 3P 030 1527 2 x12 ml prokaryotes 34 10P 034 83 2 x12 ml prokaryotes 30 3V 030 1527 1 ml viruses 34 10V 034 83 1 ml viruses 30 4P 030 1219 2 x12 ml prokaryotes 34 12P 034 17 2 x12 ml prokaryotes 30 4V 030 1219 1 ml viruses 34 12V 034 17 1 ml viruses 30 5P 030 914 2 x12 ml prokaryotes 39 2P 039 606 2 x12 ml prokaryotes 30 5V 030 914 1 ml viruses 39 2V 039 606 1
74. cted beams is a useful diagnostic of whether the system is functioning correctly 3 When Autosub was flying deep on the journey from deployment to the survey area and on the return journey the apparent draft of the ice was very large e g 100m for mission 321 From visual observations these sections of the survey were under open water with only the occasional area of ice The correlation plots show a characteristic hyperbola on most pings This may result from a single acoustic spike near the detector being interpreted as a sea surface return This is further confirmed by the fact that it is detected by amplitude on all beams Later analysis of the precise shape of the hyperbola may show that it agrees with this interpretation Prior to the deployment the ship surveyed the area for the presence of icebergs either from the ice edge using radar or by entering the ice and searching visually and using radar 113 4 The appendices at the end of this section include summary information on each mission including information taken from the linestat files The linestat files for missions 321 322 and 323 include incorrect low values for the distance travelled the error is of an order of magnitude The distance travelled for mission 324 is plausible 5 After Mission 321 it is believed that there was a probable bad connector on the transducer At this time of writing the Autosub team cannot be more specific but it is likely that it was not
75. cy is much higher than the lowest frequency in the transmitted pulse Thus we are sampling well above the nyquist and so to display must filter the data to produce something close to a power envelope otherwise the data looks hopelessly noisy I used a fifteen sample long centre weighted filter to give plot below of the first and second segments 103 First Segment 500 1000 1500 sample number 100 200 300 400 500 600 trace number Figure 15 First segment from M309 corrected for sub depth The upper lines is the transmitter pulse reflecting the changing vehicle depth and the bottom line is the seabed reflection In this plot the depth of the primary reflection is 3000 6801 samples 9801 9801 0 048 470 metres And the amplitude of the undulations in the plot are 300 0 048 14 4 metres This segment is from the Autosub in a bottom tracking holding pattern of 50 metre circles shown below Note the perfomance of Autosub in tracking the bottom It is generally a little delayed and doesn t quite get into the troughs but this is not entirely fair as a test because it s turning in circles at the same time This is as expected from the configuration of the system and control software There are 10 5 pattern repetitions in the profiler data and 10 5 circles in the track data This is good 104 M309 Legend M309profiler1st Events 7 M309profiler2nd Events End 10 58 PUT A Y 2 Stat 16 40
76. d on 19 Jul 02 G 1 04108582 001 1 38996218e 000 3 42550982 003 3 12641143 004 CTcor 3 2500e 006 CPcor 9 57000000e 008 Slope 1 00000000 Offset 0 00000 3 Frequency channel 2 Pressure Digiquartz with TC Serial number 67241 Calibrated on 30 Jun 2000 C1 4 461418e 004 C2 3 038286e 002 C3 1 224130 002 D1 3 645500e 002 D2 0 000000e 000 T1 2 999608e 001 T2 3 512191e 004 T3 3 729240e 006 T4 4 918760e 009 T5 0 000000 000 47 Slope 0 99992000 Offset 0 88150 AD590M 1 283280 002 AD590B 9 474491 000 4 Frequency channel 3 Temperature 2 Serial number 032191 Calibrated on 19 Jul 02 G 4 31967419 003 6 38837657e 004 I 2 27990979e 005 J 2 17976156e 006 FO 1000 000 Slope 1 00000000 Offset 0 0000 5 Frequency channel 4 Conductivity 2 Serial number 019112 Calibrated on 19 Jul 02 G 4 16212062 000 5 36713913e 001 I 7 86598365 004 J 6 80295512 005 CTcor 3 2500e 006 CPcor 9 57000000 008 Slope 1 00000000 Offset 0 00000 6 Voltage channel 0 Altimeter Serial number 2130 26993 Calibrated on N A Scale factor 15 000 Offset 0 000 7 Voltage channel 1 Free 8 Voltage channel 2 Oxygen SBE 43 Serial number 0245 Calibrated on 27 Aug 02 Soc 4 0080e 001 Boc 0 0000 Offset 0 4413 Tcor 0 0014 48 Pcor 1 35e 004 Tau 0 9 Voltage channel 3 Free 10 Voltage channel 4 Oxygen S
77. d then turned to the NNE to join a former ice stream described by Canals et al 2000 that flows across Bransfield Strait to the continental shelf break Fig 4 Additional swath lines were obtained from Neumayer Channel and from the NE of Gerlache Strait passing Twin Hummock Island into Bransfield Strait The section into Bransfield Strait demonstrates clearly the convergence of two sets of large scale sea floor lineations one from Gerlache Strait and another from the east Marguerite and Ryder bays adjacent to Rothera Due to high winds at Rothera we were unable to dock immediately and took the opportunity to collect swath bathymetric data in the approaches to the BAS base and the arm of Marguerite Bay to the south of it An area of about 550 km was surveyed Fig 5 Preliminary inspection of the image mosaic and sub bottom profiler records shows that the sea floor is mainly of bedrock with a strong surface return on TOPAS There are few pockets of sediment Some of the bedrock features especially in the south eastern part of the image mosaic are streamlined in a NNE SSW direction that is approximately along the long axis of the fjord We interpret the streamlining of bedrock to be a product of glacial erosion and that the forms record past ice flow from the NNE Fig 5 References Lowe A L and Anderson J B 2002 Reconstruction of the West Antarctic ice sheet in Pine Island Bay during the Last Glacial Maximum and its s
78. d035 19 03 2003 078 23 34 113 039 71 4936 113 2 34 71 29 61 550 539 9 18 84ctd036 20 03 2003 079 09 35 114 7384 71 3146 114 44 3 71 18 88 2057 2082 2 47 84ctd037 20 03 2003 079 12 25 114 4983 71 4302 114 29 9 71 25 81 1494 1467 2 3 84ctd038 20 03 2003 079 14 14 114 4873 71 4739 114 29 24 71 28 43 1027 1022 1 22 84ctd039 20 03 2003 079 16 00 114 5441 71 4982 114 32 65 71 29 89 615 606 7 89 84ctd040 20 03 2003 079 17 35 114 2854 71 5471 114 17 13 71 32 83 525 513 7 59 84ctd041 21 03 2003 080 12 28 113 4412 71 1291 113 26 47 71 7 75 2297 2325 8 06 84ctd042 23 03 2003 082 20 45 102 6363 70 6548 102 38 18 70 39 29 4100 507 999 84ctd043 24 03 2003 083 20 01 100 7157 70 5355 100 42 94 70 32 13 3401 507 999 84ctd044 25 03 2003 084 19 08 98 4754 70 4689 98 28 52 70 28 13 3579 507 999 36 This output an ascii file with the extension cnv Finally the Sea Bird Electronics Inc Data Processing software version 5 25 Cell Thermal Mass module was used to remove the conductivity cell thermal mass effects from the measured conductivity This correction followed the algorithm dt temperature previous temperature ctm 1 0 b previous ctm dedt dt and corrected conductivity c ctm and a 2 alpha sample interval beta 2 b 1 2 2a alpha dedt 0 1 1 0 006 temperature 20 with alpha set to 0 03 beta set to 7 0 This routine output a
79. direction of motion as derived from the bottom track ship s motion This was achieved using the Unix script adcp calibration exec Input files 84bot abs Output files 84bot abs 222 where a or p for am or pm 55 84adpexecO 84 Sdadp RVS data 84 start adp go 84adpexec0 1 Contains clock drift Temperature data P correction 84adp corr 84adp corr 84adp t 84adp t 84adpexec2 84ash01 int 84adpexec 1 Ashtec correction Clock drift correction 84adp true 84adp true Data suitable for calibration 84adpexec3 84adpexec3 A 1 0 0 Using calculated calibration 84adp cal variables A and if available 84adp cal 84adp cal 84adpexec4 84adp cal abnv841 av Calculate absolute velocity 84adpexec4 abnv841 av 84adp abs Calculate absolute velocity 84adp abs Sdadp abs adcp calibration exec 84adp abs Calculate A and 2 Figure 14 ADCP Processing Flow Chart 56 File variables 84adp Time bindepth Velew Velns Velvert Velerr Ampl Good 84adp t Time bindepth Velew Velns Velvert Velerr Ampl Good 84adp corr Time bindepth Velew Velns Velvert Velerr Ampl Good 84adp true Time bindepth Velew Velns Velvert Velerr Ampl Good a ghdg
80. e Figure 34 shows the same data but for only the lawnmower survey area with the scale adjusted to show plausible ice draft values 322 18 12 23 3 03 102 40 178 70738 841 8 No swath data TO TO TO 03 40 24 3 03 102 24 01 70 21 37 323 16 47 24 3 03 100 42 44 70 32 1 6 No swath data After this mission it was realised that the current settings for the swath TO TO TO system could potentially lead to data drop out Only values above a certain to be ascertained draft were measured all other values were recorded as 02 30 25 3 03 100 41 27 70 29 67 zero 324 15 32 25 3 03 98 29 7 70724 95 6 No swath data The surface offset for this mission was set at 15 metres compared with 5m TO TO TO for the previous missions This offset was hoped would capture data about any values that were below zero i e apparently above sea level 07 00 26 3 03 98 16 01 70 22 84 As estimated by bridge officers 116 Appendices Summary of Performance of Mission 321 0015 20030322 163838 raw 0015 20030322 170915 raw 0015 20030322 173951 raw 0015 20030322 181028 raw 0015 20030322 184105 raw 0015 20030322 191142 raw 0015 20030322 194218 raw 0015 20030322 201255 raw 0015 20030322 204332 raw 0015 20030322 211409 raw 0015 20030322 214446 raw 0015 20030322 221522 raw Start Posn Start Posn End Posn End Posn No of No No Valid Min Valid South West South West pings soundi
81. e Unix exec gyroexec 65 0 Reads in the gyrocompass data and removes the inevitable bad data Steps datapup transfers the data from RVS binary files to pstar binary files Pcopya resets the raw data flag on the binary file Pheadr sets up the header and data name of the file Datpik forces all the data from the gyro to be between 0 and 360 Output files 84gyr raw The script also appends the day file to the master file called S4gyr0 Electromagnetic Log The Electromagnetic Log gives water velocity relative to the ship in both the fore aft and port starboard direction This data was read in 12 hour time periods using a simple exec emlexec0 0 Reads data from the Electromagnetic Log into pstar format Steps datapup transfers the data from RVS binary files to pstar binary files Pcopya resets the raw data flag on the binary file Pheadr sets up the header and data name of the file Output files 84eml raw Doppler Log The Doppler Log gives water velocity relative to the ship in both the fore aft and port starboard direction This data was read in 12 hour time periods using dopexec0 0 Reads data from the Doppler Log into pstar format Steps datapup transfers the data from RVS binary files to pstar binary files pcopya resets the raw data flag on the binary file pheadr sets up the header and data name of the file 66 Daily navigation processi
82. e Translator version 2 90 04 was subsequently modified which enabled the data to be read by the Windows version of Sea Bird software Missions 320 to 324 were processed with the new software The final format of the data was identical with both versions of the Sea Bird processing software the only difference should be found in the header The configuration files used were named 0696jr84 con for the missions with the original conductivity sensor and 0696jr84b con for the missions with the replacement conductivity sensor Missions will contain no useful fluorometer and transmissometer data due to the absence of these sensors 82 ADCPs and Navigation Steve McPhail Configuration For JR84 Autosub was configured with an upward looking 300kHz RDI workhorse Acoustic Doppler Current Profiler ADCP and a downward looking 150 kHz ADCP Both fire through 3 mm polyethylene acoustic windows An IXSEA PHINS Fibre Optic Gyro based inertial navigation system INS is coupled to the 150 kHz ADCP sonar head within a titanium pressure case thereby maintaining fixed alignment offset between the ADCP and INS The INS position drift performance is known to be inadequate without a velocity input from the ADCPS Navigation is most accurate anticipated 0 2 of distance travelled when the downward looking ADCP bottom tracks which is usually possible at ranges up to 400m The upward looking ADCP can provide velocity aiding when tracking the underside of a
83. e depressed by ice and difficult to see 3 Surfaced 200 m clear of ice edge 3 Argos continued to come in 4 Ice surrounded sub whilst planning short under 4 Broke GPS antenna sacrificial ice mission 5 Bent port stern plane down easily repaired Mission 321 1 Launched but came back to surface 1 Stopped mission Dive weight launch in ice 2 Mission successful second attempt 2 Timer set too short Lawn mower survey then 3 Recover repack lines out at 250m 4 Adjust timer 5 No problems 6 Good data Mission 322 1 Launched but came back to surface 1 Timer too short 15 secs instead of 15 mins Repeat of above but a bit 2 Went to wrong position top circle and seemed to 2 Allowed to time out so started leg out of ice considered a lot safer than further into ice 15 NM continue drifting south and a bit east surface 3 Timed out and headed out of ice 3 Continuing with survey mission risky if southerly drift continued may 4 SeaPam not talking well no transponder and not go far enough to get out of ice navigating using up ADCP relative to poor digital the ice moving at 0 5 knts south 4 Strong current 5 Repositioned SeaPam transducer Mission 323 1 Apparently completed mission successfully 1 Sub took avoiding action on several occasions down ADCP targets Launch outside ice and 2 Swath data not looking so good wrong reset run in for survey Mission 324 24 25 0303 1 Waited at ice edge to intercept could hear it on 1
84. e periods of inconsistent ADCP data as displayed on the arrow graph could also be identified and the relevant data cycles and removed This sometimes resulted in more than one block of data cycles being copied over to the new file for averaging using allav 60 z 4257 E 52253 SOS x 24 BERS Seen SS 5 XO SS EA OES NNN SRT 5 CUN SSS XS eaten NS HH gy SS SN OON OR EYE Pies OY VD fj MALA LIA 2 A SES SYS es Zu u T TT ves RIS 22 TIL ELLE 7 21 24 Ay 2 BUC EZ WELZ gt 7227 LIA gt 8 NIA DET ELPA 315 270 180 135 1160000 1220000 0 0 00 Figure 17 Arrow graph plot of abs ADCP data for duration of CTD deployment 61 Navigation data Sarah Hardy and Mark Brandon Open University There were five navigational instruments for scientific use on the RRR James Clark Ross listed in Table 7 below Although the five instruments appear in some cases to be similar they are all unique As well as the three GPS systems listed in Table 7 there are additional GPS systems on board the JCR for the ship s use These are a Leica MX400 and two Ashtec G12 receivers as part of the dynamic positioning system In addition there is a Racal Satcom which receives GPS SV range correction data via INMARSAT B This data is passed to the Trimble Leica and G12 receivers allowing them to operate in
85. e ship s navigation report However it is well known that in addition to having an inherent error gyrocompasses can oscillate for several minutes after a turn before steadying on a new course There is also an additional deviation of the gyrocompass that varies as cosec latitude To overcome these difficulties the ADCP is corrected with data from the Ashtec ADU 2 see navigation report The Ashtec cannot be used instead of the gyrocompass because Ashtec coverage is not continuous but the data can be corrected on an ensemble by ensemble basis As a result of the standard processing as detailed in the navigation report the edited Ashtec data is held within a file as data of 53 2 minute averages This data still contains large spikes which are removed using an interactive editor Any gaps created by this editing or previously existing in the data are linearly interpolated by a further program The gyrocompass correction file 84ash01 int is then applied to the ADCP data through the Unix script 84 2 The east velocity velew and north velocity velns from the ADCP are converted to speed and direction and the heading correction as calculated from the gyrocompass correction file applied to both the gridded water track data and non gridded bottom track data The program then converts the data back to east and north velocities ready for the A and calibrations performed in the next processing step Shou
86. ed at about 1000 m Phoned Simrad provided fix Removed front 2 harnesses Ran extension from rear harness to SeaPam in front removed 2kg ballast weight Modified range and Seapam interface pots joined network before internal connector and chopped pins off which were close to chassis Tightened loose bulkhead connector on AquaLAB Moved P2 to rear Moved P1 to nose but not used Launched and left to wallow for 70 mins data looked good Mission 314 14 03 03 Run south for 15 mins at 100m turn back to WP heading down to 1320m Data after wallow test had some drop outs but acceptable Mission excecuted with no apparent problems OOo E Os UM Closer scrutiny of data revealed that it was better but not good dropouts at 700m descending and 200m ascending Analysis of data continuing Remove extension Re instate front harness connected through link from rear P1 moved to rear to join P2 Seabird in loop AquaLab in loop Range finder back in loop Damaged SeaPam serial interface pot by plugging battery power into network hopefully repaired Check individual skts on network connectors all seem OK in rear Plotting drop out counts against time and pressure indicated pressure related problem suspect bulkhead connectors REMOVED TAIL to change network bulkhead connectors on GPS had to REMOVE NOSE to free up trapped wire holding GPS unit in power
87. eet We interpret the streamlined features in the newly described trough on the western side of outer Pine Island Bay in the same way as indicating past ice stream flow The direction of streamlining is SE NW and we suggest that this may mark the former flow of an expanded Thwaites Glacier on the outer shelf given that Lowe and Anderson 2002 report S N orientated lineations from further east in the Bay which are more probably related to the past flow of a full glacial Pine Island Glacier TOPAS records from the area of lineations in the cross shelf trough show that the sea floor is sedimentary and that there are in some places reflectors at a few metres depth defining an upper unit in whose surface the lineations are formed It is noteworthy that on the steep eastern side of the cross shelf trough several reflectors appear to cross cut one another each with a similar acoustic unit above This may mark the shifting margin of the palaeo ice stream The second main feature observed on swath bathymetric images of the shelf was grooves of varying width depth and orientation The grooves were formed mainly in water depths of less than about 460 490 m Fig 3 They are similar to the scours produced by iceberg keels on many glacier influenced continental shelves and represent the irregular drift tracks and grounding of icebergs derived from Thwaites and Pine Island glaciers The floating tongues of these ice streams are known to be about 500 m th
88. em Here y axis is depth in metres x axis is tracenumber M309 M309profilertst Events M309profiler2nd Events 620 Meters 1 Start 17 30 02 Figure 29 Trackplot for the second segment of M309 107 Assessment of Simrad EM2000 multi beam sonar David Vaughan With assitance from Toby Benham James Perrett and Chris Banks The Simrad EM2000 system is installed on Autosub with an option of either up or downward configuration It was trialed in August 2002 but was at that time not producing satisfactory results although the Autosub team not being familiar with this swath bathymetry data did not identify some of the problems This lack of expertise on the trials was unfortunate but could not avoided after Lieve Vanneste resigned from her position on the Vaughan project just prior to the sea trials During the early trials in Gerlache Strait M307 the installation of the EM2000 system was steadily improved And valuable science data was acquired beneath sea ice This system is however not yet a turn key system and will require some dedicated sea trial time to reach a fully operational state Outstanding issues DSO correction It appears that in the present configuration the EM2000 will reject any echo that is apparently above the nominal sea surface calculated onboard using the vehicle depth and the sound velocity that is given to the vehicle at setup It needs to be confirmed with Simrad that this is indeed the
89. en every 30 minutes throughout the run The Autosub carries upward and downward looking ADCPs which collect data every 2 seconds throughout the mission Conductivity Temperature Depth Absolute levels of temperature and conductivity are investigated by comparing a shipboard CTD cast to an Autosub mission On mission 316 17 17 GMT 16 March 2003 71 215 113 320 the Autosub dived to 1200 m and returned On cast 84ctd011 the ship s CTD system was lowered to 2000 m 05 10 GMT 16 March 2003 71 223 113 345 The temperature conductivity and oxygen profiles in the depth range of overlap are shown The two temperature sensors on Autosub are in excellent agreement as well as the two conductivity sensors See Table 12 The shipboard sensors are also shown for comparison Both the temperature and conductivity profiles show the same broad features However the two casts show significant differences in detail particularly above 700 meters Given the separation in time and space of the two casts 12 hours and 1 6 km and the internal consistency of the two instruments these variations are acceptable Below 700 meters we expect nearly identical results since the Circumpolar Deep Water found there is a large and slowly varying water mass The differences there are very slight a maximum of 0 01 deg C and 0 001 S m Although slight these differences are significant A plot of temperature versus salinity for the two instruments shows that wa
90. ent Station 84ctd018 This cast was aborted at 40 dbar depth due to a failure in the dynamic positioning system of the ship Station 19 was at the same location 45 SBE43 Dissolved Oxygen Sensors Two new SBE43 DO sensors serial numbers 0245 and 0242 purchased during summer 2003 were used throughout the cruise Neither performed well There was always an offset of 1 ml l between the readings from the two instruments figure 13 While the overall level of the secondary sensor looked the better of the two this one also suffered markedly from pressure hysteresis with an offset of up to 0 3 ml l between downcast data when the sensor was being loaded and upcast data when it was being unloaded During stops in the upcast when niskin bottles were being fired the reading from the secondary sensor relaxed to a value intermediate between the down and up trace The primary sensor showed much less hysteresis but never recorded oxygen levels close to saturation even at the surface in open water Since no underway Winkler titrations were performed processing of the data will have to await the post cruise calibration of the sensors by Seabird Although the final absolute oxygen concentrations are likely to be subject to relatively large errors the main motivation for recording the data was to help quantify mixing within the main pycnocline For this relative changes in concentration through the water column are most important Any future users
91. essure sensor and the action of pressure on the conductivity cell The output file is of the form CCcnvNNNtm CNV A second file of the form CCctdNNN ros is also created These files were saved on the D drive of the CTD PC with a separate folder for each CTD They were then transferred to the UNIX system jruf and placed in the directory pstar data ctd ascii_files 84ctdNNN where NNN is the event number of the cast 2 To process the SBE35 data Communication must be established between the CTD PC and the SBE35 by switching on the deck unit The program used to process the data is Seabird terminal programme This is a simple terminal emulator set up to talk to the SBE35 Once you open the program the prompt is gt You can ask the SBE35 how it is by typing DS ds This stands for display status The SBE35 responds by telling you the date and time of the internal clock and how many data cycles it currently holds in memory The next thing is to click the capture toolbar button and enter a sensible filename Once done the data can be downloaded by typing dd This stands for dump data The data currently held in the memory is listed to the screen This can be slow due to the low data transfer rate Once finished downloaded one clicks on the capture button to close the open file and the clears the memory of the SBE 35 using the command samplenum 0 39 Finally one should type ds to check that the memory is clear before shutting down the
92. f files The corresponding bnv file correctly gives dates as 04 03 03 ie a day later Find corrections for the clocks File timecomp log gives corrections to clocks The Edgetech uses its internal clock Fsau f014038 which is referenced to on board pc clock Autosub9 Best navigation data from bnv is referenced to Asublogl which is referenced to Autosub9 and local Asubtosh1 I used the following line of reasoning based on the timecomp log file to establish the time correction near the start and end of the mission At start of mission time for Fsau is Time at server Fsau f014038 is 13 34 53 55 Local time is 13 34 45 351 Difference is 8 199000 seconds i e difference onAsubtosh1 At the same time Time at server is 13 34 58 13 Local time is 13 34 48 856 Difference is 9 274000 seconds And Time at server Asublog1 was 04 03 2003 02 11 31 when time at host was 04 03 2003 02 11 32 Local time is 13 34 52 671 Difference is 1 000000 seconds Difference 1 s 101 so Fsau is 8 199 s faster than local Autosub9 is 9 274 s faster that local and Asublog1 is 1 s slower than Autosub9 Therefore Asublog1 is 9 274 1 faster than local FSau is 8 199 9 274 1 faster than Autosub9 ie Nav 0 075 sec And at the end of the mission Time at server WFsau f014038 is 20 54 49 56 Local time is 20 54 40 976 Difference is 8 584000 seconds Time at server Autosub9 is 20 54 53 62 Local time is 20 54 43 589
93. g Normal Line Active Line Polygon Figure 31 Autosub EM2000 swath bathymetry data from the underside of sea ice collected during M321 111 Appendix Autosub missions Log M307 03 03 03 Gerlache Strait test Swath down Imprecise lat long position produces stepped navigation All data is smiles M308 03 03 03 Gerlache Strait test Swath down Imprecise lat long position produces stepped navigation All data is smiles M309 04 03 03 Gerlache Strait test Swath down Imprecise lat long position produces stepped navigation All data is smiles M310 08 03 03 Off Thurston Island Swath up Mostly a dive test spiral dive no profiler but swath data looks like garbage M311 09 03 03 Off Thurston Island Swath up Swath data is available only in central beams but system still appears to think that it is looking down and has added the ROV depth instead of substracting it Some of these data could be useable if corrected for upward looking and cleaned but it is only looking at the sea surface M312 10 03 03 Off Thurston Island Swath up Swath data is available but navigation appears to have gone potty Lots of broken line segments Still appears to be thinking that it s looking down M313 13 03 03 Continental shelf off PIB No swath or profiler M314 14 03 03 Continental shelf off PIB Swath up data available It s now looking up and understand that it is looking up For section at a reasonable depth m
94. had moved further west leaving a corridor only 40 miles wide between it and a line of icebergs extending north west from Thwaites Glacier Tongue A persistent southerly breeze brought cold temperatures causing new ice to grow over the entire open area to the south of the sea ice tongue While only a few inches thick it appeared to be rapidly consolidating and would already have presented a significant hazard to Autosub On 14 March Figure 1c the decision was taken to abandon further attempts to access the Bay Conditions at the other ice fronts of the Amundsen and Bellingshausen seas had not improved sufficiently to justify a long transit eastward and indeed a week later Figure 1d the new ice growth had effectively barred access to any conceivable ice shelf work site The only part of the Amundsen Sea continental shelf that was relatively free of ice was further to the west where a trough cutting the shelf break had been identified during a Nathaniel B Palmer cruise in Feb Mar 2000 This location had the potential to be of scientific interest for two of the four projects on board so was made the primary location of JR84 work Further work on the multi year sea ice north of Thurston Island was conducted on the eastward return leg of the cruise 4 ey 20 0 08 March 0 0 15 35 65 85 95 97 99100 gt 0 15 35 65 85 85 97 99100 y AU 1 A g 1 0 15 35 65 85 95 97 89100 a0 15 35 65 85 95 97 99100 i 1
95. he configuration of the ADCP The RRS James Clark Ross is fitted with an RD Instrument s 150 kHz although actually 153 6 kHz hull mounted Acoustic Doppler Current Profiler ADCP Unlike other NERC research ships the orientation of the transducer head on the JCR is offset by approximately 45 to the fore aft direction in hope that the instrument will give a better response in the main direction of motion i e fore aft To provide protection from ice the transducer is mounted in a sea chest recessed into the hull of the ship which is again different from the design of other British research ships The contents of the sea chest are isolated from the surrounding sea water by a 33mm thick window of Low Density PolyEthylene LDPE Within the sea chest the transducers are surrounded by a liquid composed of 90 de ionised water and 10 ethylene glycol The version of the firmware used by the ADCP was 17 07 and the version of RDI Data Acquisition Software DAS was 2 48 The software ran on a Pentium 2 266Mhz running DOS For JR84 the ADCP was configured to record data in 64 x 8 bins and in ensembles of 2 minute duration The blank beyond transmit was 4m which when added to the approximately 6m depth of the transducer resulted in the depth of the centre of the first bin depth being 14m In water depths of less than 500m the ADCP was operated in bottom track mode Water track mode was used in deeper water The bottom track mode was
96. her CTD section stations 009 and 010 across the continental slope Autosub was deployed at the second station M315 but the systems cut out at depth The Master was unhappy about continuing the CTD section into the ice at night so the ship headed for W113 20 where a trough cutting the shelf edge had been identified in 2000 taking a line well north of the ice edge A CTD section across the slope was started 16 Mar 2003 Overnight the five station 011 015 CTD section was completed We then returned to deep water along a track that maximised the swath coverage Autosub mission M316 encountered further problems at depth so we recovered the vehicle and headed south onto the continental shelf as far as we could while staying well clear of ice for overnight operations We located the western edge of the seabed trough near W114 15 and started a swath and CTD line stations 016 024 across it 17 Mar 2003 The transect across the trough was finished by morning and we headed for the 2000 m isobath at W114 Autosub completed another two test missions M317 8 which left the technical team feeling that most of the problems were solved for missions to a few hundred metres depth We started another CTD section stations 025 029 across the shelf break With this complete an extensive swath survey was started aimed at mapping the seaward end of the trough and the upper continental slope beyond 18 Mar 2003 In the morning the swath survey was su
97. hort period of red This was not regarded as a stable set up and so the system was generally run throughout the cruise on fixed time with the time allocated dependent on water depth This obviously necessitated manual changes to the SSU by us In deep water there is also a problem of low data density for each of the echo sounders On previous cruises JR71 and JR59 this problem was addressed by operating TOPAS in manual triggering mode This was also tried on JR84 in areas of deep water However the result appeared to adversely affect the transmission and reception of the EM120 in that the transmission pulse appeared grey implying that the EM120 was no longer under SSU control Again this was not considered very reliable and so all three echo sounder were triggered by the SSU in deep water with the EM120 and 500 generally on fixed time usage and TOPAS on calculated Recommendations The patch supplied by Kongsberg Simrad did not solve the EM120 centre track problem noted above We suggest that the SSU and software and the EM120 21 operating software need to be checked by Kongsberg Simrad to make sure they are functioning correctly SSU Settings Some examples of SSU settings that were used on JR84 are given below It should be noted that these were not the only settings used but are included here as examples It is anticipated that these will be modified between and during individual cruises Note the EK60 was off throughout JR84 and
98. hough later missions without a primer also resulted in varying sized sample volumes 3 3 03 All Primed 50ml Bag On Port Volume Bag On Port Volume 20 195 23 345 21 395 24 270 22 335 25 380 4 3 03 All Primed 50ml Bag On Port Volume Bag On Port Volume 1 14 25 2 10 15 empty 3 25 16 15 4 50 17 300 5 20 18 355 6 5 19 310 7 40 20 390 8 15 21 370 9 empty 22 340 10 20 23 310 11 20 24 340 12 not recorded 25 330 13 10 3 3 03 No prime bags flushed only Bag On Port Volume Bag On Port Volume 18 265 22 265 19 260 23 450 20 395 24 265 21 265 25 305 M312 10 3 03 No prime bags flushed only Bag On Port Volume Bag On Port Volume 17 25 22 315 18 285 23 370 19 275 24 365 20 280 25 415 21 290 Other bags below 17 were either empty of contained less than the 50ml prime water M313 13 3 03 No prime no flushing on bags 18 25 Bag On Port Volume Bag On Port Volume 17 22 270 18 23 265 19 265 24 395 20 265 25 260 21 265 79 M316 16 3 03 No prime no flushing on bags 18 25 Bag On Port Volume Bag On Port Volume 17 280 22 275 18 295 23 275 19 285 24 275 20 280 25 275 21 275 M317 amp M318 19 3 03 No prime
99. ick and to produce icebergs with deeper keels than for example the Ross Ronne and Amery ice shelves whose bergs are typically about 300 330 m in thickness TOPAS records from the shallower areas of the shelf where scouring has taken place show a very uneven sea floor related to the berms and central troughs of the scour marks No internal acoustic stratification is present in iceberg scoured areas as the ploughing action of the keels reworks the surficial sediments The Amundsen Sea Continental Shelf Break and Slope We also undertook geophysical measurements of the Pine Island Bay shelf edge and shelf continental slope shelf processes are clearly likely to influence sediment delivery to the continental slope Swath bathymetric and TOPAS data were collected along the shelf break and upper slope from 108 to 115 W a distance of about 220 km Fig 2 Two large blocks of swath bathymetric and TOPAS data were collected from the shelf break and upper slope connected by either single or double lines of data The two blocks on the shelf break and upper slope were of approximately 2 200 km and 800 km in area centred atl 13 20 W and 108 40 W respectively Fig 2 Where the shelf break was at less than about 500 m iceberg scours impinge right to the shelf edge In deeper water they are present only very occasionally and here streamlined sea floor lineations are found even close to the shelf break These lineations are less well defined
100. ience festival and a possible future DVD for schools 10 Marine Geophysical Investigations Dowdeswell J Evans Cofaigh T J Benham Scott Polar Research Institute University of Cambridge Introduction Our original intention had been to collect marine geophysical data from Pine Island Bay using the EM120 swath bathymetric system and TOPAS sub bottom profiler This work would have complimented that of the Autosub vehicle beneath Pine Island Glacier However given that sea ice conditions precluded ship operations in all but the outer shelf of Pine Island Bay our plans were reorganised The reconstruction of past ice sheet flow and the delivery of sediments to the Antarctic continental margin remained the central scientific theme of our work but the geographical focus of this now became the outer shelf of Pine Island Bay and the adjacent continental slope of the Amundsen Sea Fig 2 This area has been very little studied before Lowe and Anderson 2002 and so our work was largely breaking new ground Our marine geophysical work can be divided into several parts relating to both scientific questions and geographical locations The five areas were The outer shelf of Pine Island Bay The Amunsden Sea continental slope off Pine Island Bay The abyssal ocean plain in the Bellingshausen Sea Gerlache Strait Antarctic Peninsula Marguerite and Ryder bays adjacent to Rothera Uia Ue pom Work in areas 3 5 was underta
101. igation We need to produce an algorithm that filters out such spurious range returns not an simple task given the overriding requirement reliably to detect the seabed We had planned to use the upward looking ADCP for navigating relative to the under ice shelf surface when bottom track data is unavailable from the downwards looking ADCP This ability caused a problem in mission 322 23 3 2003 an under sea ice mission where the water was too deep for bottom tracking but the upward looking ADCP was able to track the sea ice The sea ice driven by the current and the wind was drifting at an appreciable speed 0 5 knts south hence seriously affecting the absolute navigation accuracy This problem was overcome in subsequent missions by a minor software change in the upward looking ADCP If the ice draught determined as the vehicle depth minus the ADCP upward range is less than 10m then upward tracking mode is disabled we would expect sea ice to measure less than 10m draught ice shelf ice to be more Despite the problems that the tracking off the sea ice caused to the vehicle navigation it was useful to get the opportunity to operate the vehicle in this mode 84 Mission Descriptions Miles Pebody An Autosub mission consists of a pre programmed series of navigation instructions that control the vehicle through a planned series of actions Usually beginning with a dive and ending back on the surface Autosub can be reprogrammed while st
102. ill in the water although it must be on the surface and within radio range of the support ship Consequently there may be more than one mission per deployment The course of the cruise was to begin with a number of Autosub test missions in the relative calm of Gerlache Strait Once the vehicles systems were found to be in order following work was then to take place under sea ice North of Thurston Island and then under the Pine Island Glacier On arriving at the sea ice work area a number of problems were found with the Autosub s control network which manifested at depths generally greater then approximately 500m Consequently a number of test mission were run to locate repair and retest the submarine After a number of days it had been ascertained that it was not possible to get to the Pine Island Bay and so an extended series of under sea ice survey mission was planned to take place once a safe working depth for the Autosub had been found Figure 22 Autosub 100m altitude terrain following track over Gerlache Strait Bathymetry 02 03 03 04 03 03 Buoyancy and Test Missions Gerlache Strait Missions M307 M309 On 2 March buoyancy trials were run with the vehicle being lowered into the water behind the ship This completed the next two days were spent running test missions The first was a general shake down run to make sure that all systems had survived the trip down from the UK and were working as they should 85 Problems were i
103. in difficult stations but in general give no more information than the primary sensors The output files are CCetdNNN cbottle CCctdNNN csamp CCsamNNN cdif which should be moved to the directories samples cbottle samples csamp samples cdif respectively The quality of the CTD calibration The mean difference in the 36 duplicate salinity samples was 0 00011 salinity units therefore we have to assume that the analysis of salinity samples was good Once the salinity data had been analyzed some samples were excluded from the derivation of calibration offsets because they were clearly sited in a poor calibration region i e in a 41 strong vertical gradient of salinity or because of clear contamination A list of excluded bottles is in the table 5 below and although there appears to be many they are almost all at shallow depths and within the halocline Table 5 Bottles excluded from the CTD calibration CTD station Bottles excluded 84ctd001 8 9 11 12 84ctd002 1 12 84ctd003 11 12 84 004 8 9 10 11 12 84ctd005 10 84ctd006 9 11 12 84ctd007 7 11 84 0008 5 3 11 84ctd009 9 10 11 12 84ctd010 7 11 84ctd011 9 12 84ctd012 10 12 84 0013 9 11 12 84ctd014 3 5 84ctd015 3 5 9 11 84ctd016 3 11 84ctd017 1 3 84ctd018 84ctd019 5 11 84ctd020 3 5 11 84ctd021 5 12 84ctd022 3 5 11 84ctd023 3 6 84ctd024 9 11 84ctd025 8 11 84ctd026 12 84ctd027 12 84ctd028 3 11 84ctd029 3 5 11 84ctd030 8 12 84ctd031 8 11 12 84ctd032 6
104. investigation as this type of connector is specified throughout for the network harness for the second build of Autosub Later in the cruise a similar leak of a Burton connector supplying the EM2000 transmitter caused loss of all swath data for missions 322 323 324 It was unfortunate that the multiplicity of software configuration issues which we had had with the EM2000 directed our attention away from this hardware fault The more serious problem was with the Impulse IE55 19 way connectors 12 of which distribute power and LonWorks communication network throughout the vehicle The problem became apparent in mission 312 10 3 2003 where the vehicle reached a depth of 1300 m and then aborted the mission due to the release nodes detecting that there had been a network continuity failure Analysis of previous mission data revealed that the fault had been intermittently occurring since the start of the cruise The fault was manifest by intermittent network data dropouts during descent at around 400 to 1000m depth and then the problem re occurred when the vehicle returned to around 300 m depth Dropouts were generally worse on ascent than descent lasted longer and occurred at a lower depth value During the next seven days we made extensive attempts to isolate the fault including checking the resistance of all the network connections in the vehicle and as a precaution replacing internal network connectors where the connector resistance was higher
105. ird ASCII format to pstar The output files are CCctdNNN raw and CectdNNN The raw file should be moved to the directory raw and the other to the directory rough 84seactd1 This exec requires the SBE35 data to have been transferred and downloaded and the salinity data to have been transferred as described above This exec produces four files CCctdNNN bottle containing the CTD data at the bottle firing points CCtdNNN samp containing the above file with the addition of the bottle salinity data and the SBE35 data CCsamNNN diff containing some residuals from the above file sampNNN bot containing salinity data from the spreadsheet in a pstar file This exec uses the CCctdNNN samp file to derive the conductivity of the salinity samples mlistis used to produce a quick and dirty plot of botcond vs deltaC A plot of bottles over the salinity profile of the CTD is then produced These plots will be produced both on the screen and printer as they are only rough plots the cast number should immediately be written on for future reference The output file is CCctdNNN cond containing the conductivity variable deltaC 40 After running the exec the files should be moved to the directories samples bottle samples samp samples diff samples salts respectively 84seactd2 This exec plots out the salinity profiles of the CTD stations and overlays the bottles on top of the profiles Obvious bad salinity samples can be spotted very rapidly
106. ken on an opportunistic basis mainly on passage to Pine Island Bay or during bad weather when other scientific operations were suspended Outer Shelf of Pine Island Bay Our survey area penetrated up to 200 km inshore of the shelf break between 108 and 114 W Figs 2 3 The major topographic feature identified for the first time was a cross shelf trough trending SE NW up to about 100 m deeper than the surrounding shelf The trough had a well defined eastern margin and a less steep western side It was about 50 km wide and approximately 600 m at it deepest This trough is to the west of the main areas covered by the earlier marine geophysical work of Lowe and Anderson 2002 who identified a mid and inner shelf trough further east and south in front of Pine Island Glacier We refer to it as Thwaites Trough Fig 3 Two main types of bedform were observed on the shelf Based on TOPAS records these appear to be composed of soft sediments that are thin and discontinuous laterally Streamlined elongate bedforms were observed in the cross shelf trough Fig 3 These bedforms were most clearly defined on the innermost part of the shelf we examined but appeared in more subdued form on the outermost shelf too These features are interpreted as glacial lineations Stokes and Clark 1999 and are inferred to be a 11 product of soft sediment deformation beneath former ice streams draining large interior basins within the Antarctic Ice Sh
107. ld there be no Ashtec correction to be made this exec can be replaced by one that adds a dummy zero value correction variable a ghdg or subsequent processing steps can be modified to omit this variable Input files 84adp corr 84bot corr 84ash01 int Output files 84adp true 84bot true Step 5 Calibration of the ADCP data A final correction is now required to correct for the misalignment between direction as defined by the Ashtec ADU 2 antenna array and the actual direction of the ADCP transducers This correction is called the heading misalignment 2 There is also an inherent scaling factor A associated with the ADCP by which the water velocities must be multiplied to scale them correctly The method of calculating A and 15 described in Box 1 These calculated corrections were then applied to both water track and bottom track velocity data through the Unix script 84adpexec3 The calibration values used during JR84 were A 1 0284 and 1 68 Input files 84adp true 84bot true Output files 84adp cal 84bot cal Step 6 The data now contains calibrated water velocity relative to the ship To derive absolute velocity the files are merged with position form the bestnav navigation file see navigation report and derive ship velocity between ensembles This velocity is then removed from the water velocity data to give absolute water velocity This is performed using the Unix scrip
108. low This particular run shows a smooth oxygen trace but later runs show more structure m323 for example The transmissometer seems to be malfunctioning with a constant value of 0 15 transmission The fluorometer shows a negligible depth dependence a noisy trace between zero and 0 03 mg m while the ship s fluorometer shows a range of 0 14 to 0 09 mg m over the same depth range One of the these was destroyed during recovery on XX March Platform Mean T2 T1 St dev T2 T1 Mean C2 C1 St dev C2 C1 Autosub 1 34 10 5 48 107 1 24x107 0 533x107 Ship 6 46 10 3 81x10 1 46x10 0 400 107 Table 12 Comparison of sensor pairs 16 March 2003 Autosub mission 316 and CTD 011 Water Sampler The water sampler presented serious problems Primarily the sample bags contained unexpected volumes of water The fact that sample bags returned with variable amounts of water suggests the possibility that leakage was contaminating the samples Several changes were made in an attempt to solve this problem See Autosub instrument section Below is a table of sample volumes and salinities for five runs Actual sample salinity was calculated using the shipboard salinometer Samples were drained from their plastic bag directly into new sterile glass bottles and the volume of the remaining water was measured CTD salinities are based on a 15 minute averages from the Autosub CTD system Unfortunately the sample
109. ly in order to produce valid coastline files we were advised to use the following Mcoast lt input text file gt f lt output text file gt r 0 Neither of these variations on the standard commands is properly described and I would have not discovered them if they had not been pointed out Finally I note that processing the EM2000 data from Autosub may never be entirely possible using the GUI menu driven Neptune system alone I recommend that anyone attempting processing make themselves aware of the commands that can only be issued from the command line Since there are three distinct sources of information the printed manual the assorted html help files and the command line instructions accessed via the h option It is not clear to me how one goes about becoming familiar with these commands without a great deal of personal investigation 7 recommend that Simrad be informed of our dissatisfaction with their manuals and be requested to rationalise these sources of information Exporting grids Kongsberg helpdesk have admitted that no facility is available to export gridded data except as a list of xyz points This is not satisfactorily in that it means that data would need to be exported from Neptune and then re gridded in whichever package one was hoping to use it This is very poor practice and should be avoided if at all possible We should request that Simrad supply a routine for outputting the grids in any of the standa
110. marked change from 1027 5 to 1026 9kg m at the same depth This change in density equates to approximately 1kg change in vehicle buoyancy and should not have been particularly noticeable The momentary state of neutral of buoyancy suggests the vehicle was marginally buoyant at the start and or there are parts of the vehicle that significantly compress with pressure However the vehicle had successfully dived to 1320m without any undue change needed in pitch or stern plane angle to maintain control suggesting no undue compression was happening Fig 18 The problem highlights the problems of running without an emergency abort weight the vehicle should be reweighed back at SOC to determine if there were any errors made during the final ballast and trim measurements M312 10 03 03 1400 1200 1000 Pitch deg x 10 M o Ho 800 Prop eMe bee Splane deg x10 L 15 600 A Virtual neutral buoyancy 400 200 Depth Prop speed Stern plane Pitch 200 ElapsedHours Figure 18 Mission 312 prior to adding more buoyancy 72 Date of Changes Net After Remarks change made all Buoyancy Mission ballast wts are change No lead kg 03 03 03 None None None First dip in water at Gerlache Strait to ensure it floats sub left attached to winch lines 06 03 03 1 1 kg added Weight added to compensate
111. ml viruses 30 6P 030 605 2 x12 ml prokaryotes 39 4P 039 508 2 x12 ml prokaryotes 30 6V 030 605 1 ml viruses 39 4V 039 508 1 ml viruses 30 7P 030 454 2 x12 ml prokaryotes 39 6P 039 305 2 x12 ml prokaryotes 30 7V 030 454 1 ml viruses 39 6V 039 305 1 ml viruses 30 8P 030 302 2 x12 ml prokaryotes 39 8P 039 201 2 x12 ml prokaryotes 30 8V 030 302 1 ml viruses 39 8V 039 201 1 ml viruses 30 9P 030 204 2 x12 ml prokaryotes 39 10P 039 102 2 x12 ml prokaryotes 30 9V 030 204 1 ml viruses 39 10V 039 102 1 ml viruses 30 10P 030 104 2 x12 ml prokaryotes 39 12P 039 27 2 x12 ml prokaryotes 30 10V 030 104 1 ml viruses 39 12V 039 27 1 ml viruses 30 11P 030 54 2 x12 ml prokaryotes 42 2 042 508 2 x12 ml prokaryotes 30 11V 030 54 1 ml viruses 42 2V 042 508 1 ml viruses 30 12P 030 28 2 x12 ml prokaryotes 42 4P 042 406 2 x12 ml prokaryotes 30 12V 030 28 1 ml viruses 42 4V 042 406 1 ml viruses 34 2P 034 631 2 x12 ml prokaryotes 42 6P 042 303 2 x12 ml prokaryotes 34 2V 034 631 1 ml viruses 42 6V 042 303 1 ml viruses 34 4P 034 457 2 x12 ml prokaryotes 42 8P 042 203 2 x12 ml prokaryotes 34 4V 034 457 1 ml viruses 42 8V 042 203 1 ml viruses 34 6P 034 406 2 x12 ml prokaryotes 42 10P 042 102 2 x12 ml prokaryotes 34 6V 034 406 1 ml viruses 42 10V 042 102 1 ml viruses 34 8P 034 305 2 x12 ml prokaryotes 42 12P 042 16 2 x12 ml prokaryotes 34 8V 034 305 1 ml viruses 42 12V 042 16 1 ml viruses Autosub Operations Nick Millard Steve Mc
112. mperature correction 84 0 1 performs a correction on the water bottom track velocity data due to the presence of the de ionised water ethylene glycol mix within the sea chest This correction was derived by Mike Meredith BAS and Brian King SOC The following text is Dr Meredith s description of the steps involved The ADCP DAS software assumes that the fluid surrounding the transducers is ambient seawater and derives a speed of sound through measured temperature at the transducer head and an assumed salinity of 35 However a correction is clearly needed to account for the fluid being the 90 de ionised water 10 ethylene glycol mixture instead of seawater From point measurements obtained from RDI we previously derived the following equation for the speed of sound through the mixture as a function of temperature 1484 3 60951 0 03522 The individual velocity measurements from which this equation was derived to an accuracy of 0 01 with the environmental conditions being known to within 35kPa pressure and 0 5 C temperature was used to derive a correction term to adjust the speed of sound assumed by the DAS to one appropriate for the mixture in the sea chest The correction term was 1484 3 6095t 0 0352 1449 2 4 6t 0 055 0 00029t 52 This correction is applied to both the raw water and bottom tracked velocities using the Unix script 84 0 1 A further correc
113. n Poor weather overnight made for slow progress 21 Mar 2003 The eastern CTD section was extended by one station 041 during the morning before a strengthening of the wind ruled out further work Since the swath was producing poor data the decision was taken to begin heading east to find multi year ice for Autosub thickness mapping runs We followed the shelf edge then ran parallel to our outgoing track to maximise the benefit of the underway swath mapping 22 Mar 2003 At first light the ship turned south to the ice edge to find a suitable site for an Autosub mission We entered the ice and deployed the vehicle on a mission M321 at S70 58 W105 44 to map the underside of the ice over a box then dive deep and head for open water Ship and Autosub rendezvoused in open water in the evening and we continued further east overnight 23 Mar 2003 A repeat of yesterday We entered the ice to deploy Autosub M322 at S70 39 W102 40 in the morning then followed it out to rendezvous in open water in the evening After the deployment a CTD station 042 was run to 500 m depth The ice thickness mapping section of the mission was aborted as a result of navigational errors in the early stages We made further easterly progress overnight 24 Mar 2003 At first light the ship began the usual search for a suitable site for a sub ice mission site ice edge at S70 32 W100 43 was tighter so the decision was taken to launch A
114. n dat an ascii configuration file containing calibration information with the extension con an ascii header file containing just the sensor information with the extension hdr and an ascii file containing the data cycle numbers at which a bottle was closed on the rosette with extension bl After the CTD the data were converted to ascii engineering units by running the Sea Bird Electronics Inc Data Processing software version 5 25 Data Conversion module The full data processing path is described below The calibration for each sensor was as follows For the Pressure Sensor r cp 51 1 Where P is the pressure T is the pressure period in uS D is given by D D DU U is the temperature in degrees centigrade T is give by T TU C is C C CU CUW all other coefficients are listed in the Appendix For the Conductivity Sensor jf Where the coefficients are given in Appendix 6 CTcorr and Cpcorr is pressure and temperature For the Temperature sensor Temp ITS 90 l 27315 Where all of the coefficients are given in the Appendix and fis the frequency output by the sensor 34 Table 2 Full details of CTD measurements taken on JR84 STATION YYYY MM Day of HH MM Decimal Long decimal Lat Long long Lat lat min uncorr ctd Max dpth DD Year deg min
115. n No of No No Valid Max Depth Min Depth Mean Depth Valid South West South West pings soundings soundings m m m soundings 7 015 20030324 161200 RAW S70 32 140 1W100 42 467 S70 32 121 100 42 450 1796 199356 8113 85 13 0 01 23 36 0015 20030325 012301 RAW IS70729 999 100 42 783 S70 30 490 100 43 245 1802 200022 6551 97 38 oot 7 04 0015 20030325 005224 RAW S70 30 756 1 100 43 415 S70 29 999 W100 42 783 1807 200577 88132 139 65 47 85 79 58 43 94 0015 20030325 002148 RAW S70 32 188 1 100 44 579 570 30 756 100 43 416 1793 199023 88115 142 31 62 58 80 31 44 27 0015 20030324 235111 RAW S70 33 619 1 100 45 750 S70 32 188 W100 44 579 1800 199800 88171 142 77 61 27 80 35 44 13 0015 20030324 232034 RAW S70 35 056 W100 46 923 S70 33 619 W100 45 750 1799 199689 87897 144 12 60 41 Fos 44 02 0015 20030324 214844 RAW S70 37 084 _ w100 47 926 570 37 166 __ 100 46 629 1793 199023 2248 32 62 0015 20030324 204731 RAW S70 37 305 100 47 057 S70 36 980 1 100 47 680 1801 199911 26075 287 96 0 01 164 9 13 04 0015 20030324 224958 RAW S70 36 528 W100 47 579 S70 35 057 W100 46 924 1802 200022 80665 144 48 60 03 802 __ 40 33 015 20030324 201654 570 37 351 1 100246 713 570 37 304 W100 47 057 1794 199134 63835 289 42 177 7 32 06 0015_20030324_221921_RAW S70 37 166 100 46 629 570 36 529 100 47 579 1801 199911 35970 275 46 bot 1 17 99
116. nce we concluded that this was not a sound velocity problem Correspondence received from Kongsberg Simrad during the cruise Kjell Nielsen product manager Multibeam KS stated that this centre track effect was due to a timing problem with the software upgrade installed in October 2002 and related to the I degree receiver He stated that the problem had now been fixed A patch was subsequently e mailed to us on the ship and Jeremy Robst BAS ITS installed it However the problem has since occurred again in deep water It should be noted that the problem first occurred during the initial transit south across Drake Passage from Stanley during a period when we were not running TOPAS Through adjustment of the SSU settings principally by increasing the amount of time given to the EM120 to transmit and receive using the Fixed Time option under Time Usage we found that the problem could be alleviated Throughout the cruise it was generally found that using the Fixed Time option gave a better less anomalies record on the EM120 We did use Calculated Time for both the EM120 and EA500 on several occasions In deep water when the centre track anomaly occurred it was notable that this did not solve the problem Furthermore this also appeared to adversely affect the trigger pulse of the EM120 on the SSU display whereby the transmission pulse turned grey instead of red followed by a period of red then green and then an additional s
117. ne channels perhaps linked to turbidity current activity These acoustic facies and the sediments they describe from our two sections across Charcot Canyon are presumably derived from mass wasting processes on the Bellingshausen Sea continental slope to the south Gerlache Strait During the initial testing phase of the Autosub AUV there was an opportunity to run a number of swath and TOPAS lines overnight in the Gerlache Strait Antarctic Peninsula The swath bathymetric image mosaic produced covers an area of about 650 km Fig 4 Streamlined bedforms presumably related to past glacial action are mapped TOPAS sub bottom profiler records document two types of sea floor return The first is a highly reflective and irregular surface with little or no acoustic penetration This predominates over the bulk of the sea floor in Gerlache Strait This suggests that the streamlined bedforms observed on swath records are predominantly composed of bedrock A second diffuse acoustic return is also found in some areas of the sea floor especially in the region of fluting and lineations It indicates that these features are formed in a thin laterally discontinuous unit of sediment The bedforms 14 are draped for the most part by a thin cover lt 2 m of postglacial glacimarine sediment Using these streamlined bedrock and sedimentary sea floor features as indictators of ice flow direction it appears that ice flowed from the SW into Gerlache Strait an
118. ne software In general the EM120 worked reasonably well throughout the cruise especially in shallow water lt 1000 m on the continental shelf Poor returns or an inability of the system to find the seabed occurred from time to time usually during bad weather In these situations the most useful technique was to use the Force Depth command with a depth slightly less than the true seabed Two other methods were also found to be useful varying the beam angles by generally reducing them to a narrow beam width and then once the seabed was found gradually increasing them again and secondly restricting the maximum and minimum depths to a tight range around the true seabed depth Problems These can be broadly grouped into 1 Continuous poor returns or drop outs on the EM120 when the wind direction is onto the beam This problem was the subject of much discussion during the cruise and some effort was spent near Rothera research station trying to determine the conditions under which it occurred This is documented separately see section Swath Trials by Doug Willis 2 Synchronisation of EM120 EA500 and TOPAS via the Simrad Synchronisation Unit Both COC and JEV spent much time during the cruise working on this and testing various settings and what follows is a summary of problems and recommendations based on our experience It should be noted that this summary also discusses TOPAS as the problems appear to collectively relate
119. ned file data0001177 jsf Start 2003 3 3 16 40 47 End 2003 3 3 16 50 4 900001 001 recordnum 302 1 84 secs per sample os No data lost here Opened file data0001178 jsf 100 Start 2003 3 3 16 50 5 100001 001 End 200333170 51 recordnum 301 2 146 secs per sample No data lost here Opened file data0001179 jsf Start 2003 3 3 17 0 53 End 2003 3 3 17 1 1 000008e 000 recordnum 5 1 6 secs sample 29 4 sec data lost here Opened file data0001180 jsf Start 2003 3 3 17 30 5 End 20033317408 recordnum 302 1 99 secs per sample A No data lost here Opened file data0001181 jsf Start 2003 3 3 17 409 End 20033317 503 recordnum 298 1 99 secs per sample Sgyreader ended Note the times given here are referenced to the Edgetech clock Fsau f014038 This is not the same clock as is used for navigation system or swath system and needs to be corrected see below This analysis shows that there are two separate continuous segments data0001177 79 jsf and data0001180 81 jsf with a data gap between These segments will be handled separately below Some gaps occur because the Edgetech FS AU has to be disabled whenever acoustic telemetry between the ship and Autosub is to be activiated In this case M309 the FS AU was almost certainly switched off before the in order to tell the Autosub to leave its holding pattern Also there appears to be a date error on the js
120. ng As stated above the data were read in as twice daily 12 hour files the time periods being either from 00 00Z to 11 59Z or 12 00Z to 23 59Z Our primary navigation data were taken from the RVS file bestnav This program uses the navigation data from various streams to construct a file with 30 second fixes For JR84 the primary input to bestnav was the Trimble 4000 DGPS This navigation file was read into a pstar file using the script navexec0 Reads in data from the bestnav stream into pstar format Steps datapik2 the data from RVS binary files to pstar binary iles pcopya resets the raw data flag on the binary file pheadr sets up the header and data name of the file posspd here we calculate the east and north velocities from position and time papend Output file is added to the master file pdist recalculates the distance run variable pcopya takes out the RVS calculated distance run Ouput files abnv841 The output master file abnv841 is used for all pstar required navigation information e g ADCP processing The processed data were then averaged and filtered using navexecl Navexec1 Averages and filters navigation data Steps pcopya copies output file from 0 abnv841 and changes data name pmdian removes spikes in velocity data pintrp interprets and replaces missing velocity data pfiltr data smoothed using top hat Output files abnv841 av 67 Microbiologic
121. ngs soundings Depth soundings m 70 58 069 W105 44 538 570 58 134 W105 44 457 1789 198579 1350 70 58 133 W105 44 457 570 58 054 W105 44 623 1809 200799 98632 49 12 x S x 100 3 47 65 oper ree pen i 70 50 507 W105 54 334 S70 49 700 W105 55 382 1808 200688 118823 200 86 105 6 59 21 Number of times pulse read 5 Number of times pulse read multiplied by number of beams used 117 Summary of Performance of Mission 322 Line Start Posn Start Posn End Posn End Posn No of No No Valid Max Depth MinDepth Mean Depth Valid south West South West pings soundings soundings m m m soundings 0015 20030323 171149 raw S70 38 529 102 38 945 S70 38 380 W102 39 166 1635 181485 0 0 00 0015 20030323 174226 raw S70738 380 W102 39 166 S70 38 519 _ W102 38 675 1806 200466 17376 135 31 0 02 64 05 8 67 0015_20030323_181303_ 570 38 519 102 38 675 S70 38 503 102238 516 1795 199245 73713 147 61 59 51 79 98 37 00 0015 20030323 184340 570 38 503 _ 102 38 516 S70 38 519 _ 102 38 340 1802 200022 72228 140 84 60 71 80 09 36 11 0015 20030323 191416 raw S70738 519 W102 38 340 S70 38 490 _W102 38 365 1804 200244 72597 140 02 60 86 80 04 36 25 0015 20030323 194453 570238 490 W102 38 365 S70 37 768 102238 113 1795 199245 75558 138 52
122. nitially encountered with the acoustic communications system at the ship end but generally all systems performed as expected 08 03 03 10 03 03 Open Water Test Missions N of Thurston Island sea ice Missions M310 312 Mission 310 was to test an Autosub under ice mission template and also to test the vehicle in deeper waters However it ended with a failure of one of Autosub s main 48V power regulators Once repairs had been made mission 311 was a rerun of mission 310 with an added dive to 1000m and an extra 20 minute run at 100m was added to collect upwards looking EM200 swath data This mission was ended by command to surface after 15 minutes of the swath data collection leg It transpired that the Autosub control communications network had been broken for periods during the deep part of the mission with one of the events long enough to trigger an emergency abort The fact that the vehicle had not immediately surfaced was because the mission controller was not able to receive the abort command over the broken network Mission 312 tested the sub to 1320m and the control network problem recurred 13 03 03 20 03 03 Open Water Test Mission 113 West Shelf Break Missions M313 M320 While the ship continued other work the root of the Autosub control network problems was searched for Missions 313 315 were run with different control system configurations to try and isolate the fault and all resulted in the same symptoms the vehicle c
123. no flushing on bags 18 25 Bag On Port Volume Bag On Port Volume 17 145 22 175 18 275 23 175 19 270 24 175 20 265 25 170 21 210 Conclusions When operated on deck the AquaLab operated faultlessly on all sampling programs including the extraction or primed water and flushing of the target bag However despite careful arrangement of the bags using string lines to ensure that the pipes and the bags were not kinked or folded the samples still came back after the mission with varying volumes of water This would suggest that there is either a problem related to the instrument being immersed in water or a problem when operating at depth or both A thorough examination of the AquaLab is required to check the integrity and operation of the rotary valve seals It is suggested that any laboratory tests of the instrument need to be done in water and at pressures likely to be encountered on Autosub deployments 80 Autosub SBE9 CTD James Perrett Configuration The instrument was a standard SBE9 instrument mounted in a titanium case which also incorporated a Burton 8 way connector at one end to connect to the standard Autosub wiring harness A standard Autosub LONWorks controller was also mounted in the case The initial missions 307 309 were run with temperature sensors serial numbers 2342 and 2912 together with conductivity sensors 2730 and 2760 Conductivity sensor 2730 was found to be giving different values from both 2760 and the
124. ns using the Autosub AUV Principal investigator Julian Dowdeswell Scott Polar Research Institute University of Cambridge Sea ice thickness distribution in the Bellingshausen Sea Principal investigator Mark Brandon Open University The planned activities for the cruise included Autosub missions beneath Pine Island Glacier neighbouring glacier tongues and the multi year sea ice that was expected to be present to the west of the bay Planned activities for the ship included CTD measurements and swath mapping in Pine Island Bay and the neighbouring continental shelf as well as measurements on ice floes and the deployment of three Argos beacons in the multi year pack In case of sea ice conditions in the Amundsen Sea barring access to Pine Island Bay most of these activities could have been moved to either Ronne Entrance or Marguerite Bay with George VI Ice Shelf becoming the main focus for Autosub work Failing this other alternatives including Larsen Ice Shelf had been identified In the event sea ice prevented the ship getting within 100 miles of any ice front so the only under ice Autosub missions were run beneath multi year sea ice to the north of Thurston Island Technical problems with the vehicle meant that these were somewhat limited in scope and extent Ice floe sampling and the deployment of the drifters were also completed in the same area The remaining shipboard activities focussed on a seabed trough that cut the Amund
125. nuously during cruise JR84 Exceptions were when the AUV was close to the ship and the system was turned off to reduce interference with the telemetry to the vehicle EA500 data were logged by SCS into the simulated level data stream SIM500 and retrieved into twice daily Pstar files using the script 3x84 sim This ran the Pstar routine datapup taking the jday and am or pm as the requisite inputs This data stream features uncorrected depth i e it produces bottom depth calculated assuming a mean vertical sound velocity of 1500 m The unix script then ran pedita on the uncorrected depths to remove spurious zeroes and replace them with absent data markers Since the data are often very spiky pmdian was run from whereby each successive value was replaced with the median of a moving window of five adjacent data cycles equivalent to a window of 2 minutes 30 seconds Navigation data were then merged in from pstar bestnav data set see navigation report Finally corrected depths were calculated using pcarter which feeds the ship s position into a set of Carter reference tables to correct for the assumption that vertical sound velocity averages to 1500 ms The output files created by 18 4 sim were 84sim jday a p raw containing the raw data from the SCS 84sim jday a p containing the cleaned data 84sim jday a p mrg the cleaned data plus merged navigation and 84sim jday a p corr the above data corrected using a more represent
126. o magnet a more reliable provision for a passive fail safe link should be investigated This could be a time or pressure based The 400N holding force magnet on the weight had caused problems on the Terschelling trials with ship movement causing the weight to wobble and drop It was thought with the JCR being more stable and care taken to suspend the weight accurately about its centre of gravity it would not be problem for the campaign However a weight was lost as Autosub was rotated in the gantry the backlash in the rotary head and stop start nature of the hydraulic valve was enough to shake the weight off The solution for the campaign was to take extreme care in driving the gantry smoothly For the future a latched system needs to be developed along similar line to the latched abort weight 74 Edgetech FS AU Sub Bottom Profiler James Perrett The Edgetech FS AU is a sub bottom profiler that transmits a swept frequency tone or chirp containing frequencies between 4 and 12kHz and listens for the return It can determine information about the seabed and the layers just under the seabed from the characteristics of the return echoes On Autosub the instrument is triggered by a controller connected to the vehicle s LONWorks network This controls the pulse rate and also allows the trigger pulse to be synchronised with other systems in order to control interactions between instruments The FS AU has been shown to affect acoustic comm
127. ocessing software Problems See section pertaining to the SSU under EM120 23 EPC Chart Recorder Colm O Cofaigh Jeffrey Evans TOPAS input to the EPC chart recorder was on Channel A The settings used were 0 5 second sweep 0 delay threshold 1 3 of a turn clockwise from the minimum setting trigger level 0 gain generally from 6 to 10 sweep direction from left to right print polarity centre setting Chart settings scale lines on take up on mark annotate off centre setting chart drive internal centre setting LPI varied from 75 100 contrast centre setting A ten minute time mark was supplied from the radiocode clock at the aft end of the UIC room The EPC chart recorder worked without any problems throughout the cruise until JD88 Rothera when it stopped working abruptly This followed a short period when it had been turned off but no reason could be ascertained as to why failure had occurred No spares were available on the ship during this cruise and therefore the EPC chart recorder could not be repaired BAS ETS personnel Pat Cooper said further information and spares from the supplier are required in order to fix it XBT s Colm O Cofaigh Jeffrey Evans Based on our experience of processing EM120 data it is very desirable that the correct sound velocity data are applied to the survey during acquisition and logging This alleviates the necessity of having to correct for this during post processing in Nep
128. ock correction and interpolate depth Re reading data using segyreader m but using the new time corrections gave Output from SGYreader KKK dee dee fee dee fee dee fee ee dee ee kk kk kk dee kk kk Opened file data0001177 jsf data0001 177 jsf 2003 3 4 16 40 4 450001e 001 Incorrect read of header data0001 177 jsf 2003 3 4 16 50 4 650001e 001 recordnum 302 Closed file Opened file data0001178 jsf data0001178 jsf 2003 3 4 16 50 4 850001e 001 Incorrect number for bytesToFollow 706394123 data0001 178 jsf 2003 3 4 17 0 4 850000e 001 recordnum 301 Closed file Opened file data0001179 jsf data0001 179 jsf 2003 3 4 17 0 5 050000e 001 Incorrect read of header data0001 179 jsf 2003 3 4 17 0 5 850001e 001 recordnum 5 Closed file Sgyreader ended Note that times are now 2 5 seconds less than on the previous read Now I added the depth correction using Depthinterp and M309 bnv Noting that for these files the sample interval is 64 msec this can be read from the header Which implies gt 64e 6 1500 2 0 0480 metres per sample gt 1 metre 20 8 samples So for every metre of vehicle depth below datum set in depthinterp m as the minimum vehicle depth in the segment we need to add 20 8 samples at the beginning of the trace Note that min depth in the segment must be added to get the absolute depth Display I used imagedisp m to display the depth corrected data Note that the sampling frequen
129. of the sensors who require accurate absolute concentrations are advised to ensure that there is a Winkler titration system available on board There is a bug in the version of SeaSave Win32 that was used While it is apparently possible to enter two sets of SBE43 calibration coefficients only one set is saved and then applied to both sensors The only way around this is to edit the configuration file manually with an ascii editor Once this is done care should be taken not to resave the configuration file from SeaSave otherwise the secondary sensor calibration coefficients will be overwritten 500 1000 Pressure dbar N ua e N a e o 3000 3500 Primary sensor Secondary sensor 4000 L 1 3 4 5 6 7 8 9 Dissolved oxygen ml l Figure 13 Dissolved oxygen profiles recorded at CTD station 003 46 Appendix Calibration data Configuration report for SBE 911 917 plus CTD from JR84 con Frequency channels suppressed 0 Voltage words suppressed 20 Computer interface RS 232C Scans to average s Surface PAR voltage added No NMEA position data added No Scan time added No 1 Frequency channel 0 Temperature Serial number 032366 Calibrated on 19 Jul 02 G 4 31950826e 003 6 43754128e 004 I 2 32220252 005 J 2 19161783e 006 FO 1000 000 Slope 1 00000000 Offset 0 0000 2 Frequency channel 1 Conductivity Serial number 042289 Calibrate
130. on was located and the data cycle immediately preceding this was noted as this represented the last data cycle of the last 2 minute ADCP ensemble of the stationary period of the ship 59 bindepth This block of data cycles was then copied to a new file using pcopya for further processing using allav which averaged the data cycles over one ensemble i e 2 minutes and 64 data cycles The resulting file was then viewed on an arrow graph to provide a time averaged view of the on station ADCP data see Figure 16 This process was repeated for each CTD station identified in Figure 15 84adp072 BX file 0 100 150 200 250 300 350 400 450 500 550 5250 5275 5300 5325 5350 5375 5400 5425 5450 distrun km start 1 72 210023 M cm s stop 64 72 1210021 0 0 00N 0 0 00 Figure 16 Arrow graph plot of time averaged on station ADCP data As well as listing the data cycles using mlist an additional arrow graph plot of the original abs file ADCP data for the approximate stationary period of the ship see Figure 17 aided the identification of the relevant data cycles The arrow graph plot proved useful if the ve and vn values of the data cycles for the start and stop times previously noted did not appear to fall particularly close to 0 If the ADCP arrow plot displayed consistent ADCP data within this time period then the data cycles could be selected with greater confidence of providing accurate ADCP data Any tim
131. ontinental shelf of the Amundsen Sea When it reaches the floating ice masses to the south uCDW causes the highest melt rates observed in Antarctica In total 44 CTD stations were occupied These were arranged in five sections aligned perpendicular to the continental slope and one section parallel to the shelf break approximately 10 km onto the shelf Most of the work focussed on a seabed trough that cut the shelf break near 113 5 W The trough had been identified during a Feb Mar 2000 Nathaniel B Palmer cruise to the Amundsen Sea The studies this year clearly delineated a warm tongue extending along the eastern side of the trough Temperatures in excess of 1 2 C within this core are the highest yet observed on the Amundsen Sea shelf In addition a few isolated CTD stations were occupied at the sites of Autosub deployments Data from casts 2 to 44 are shown in figure 7 and the locations of the stations in figure 8 Station 1 was a test cast in Gerlache Strait Examples of sections along and across the shelf break are shown in figure 9 Potential Temperature deg C 2 33 4 33 6 33 8 34 34 2 34 4 34 6 34 8 Salinity psu Figure 7 Scatter plot of potential temperature versus salinity data 2 dbar averages from CTD stations 2 to 44 The solid black line indicates the surface freezing point the labelled dashed lines are isopycnals referenced to surface pressure 30 Figure 8 Locations of CTD stations occupied during JR84 The lower
132. ontrol network communication breaking down intermittently when the vehicle was at depth Finally when mission 316 ended and the Autosub was recovered to the ship the problem was still evident and was traced to a faulty IE55 bulkhead connector It was evident after retest missions 316 and 317 that a number of connectors of the same design were exhibiting similar failure modes Because of this it was decided to restrict the maximum depths of the remaining Autosub missions Mission 318 was a relatively shallow 100m mission to collect EM200 swath data and to demonstrate the control network stability Deeper depths were then attempted in mission 319 first 150m and then 250m These depths were considered adequate for under sea ice survey missions and the Autosub completed them successfully Unfortunately as a result of severe weather the Autosub was damaged on recovery to the ship after mission 319 Subsequent repair took 2 days 86 Autosub Mission 321 Navigation Recovery position Ice edge 70 82 70 84 70 86 70 88 70 9 Latitude 70 92 70 94 70 96 70 98 71 105 95 105 9 105 85 1058 105 75 1057 105 65 Longitude Figure 23 Mission track of M321 20 03 03 25 03 03 Open Water Test and Under Sea Ice Surveys N of Thurston Island sea ice Missions M320 M324 Mission 320 was a short mission with a dive to 300m to test the acoustic telemetry system on board
133. ost of the data is deleted because it is above the sea surface lt 0 Where the sub dives the range becomes constant and is not tracking the sea surface M315 16 03 03 Continental shelf off PIB Swath up data available It s now looking up and understand that it is looking up For the section at a reasonable depth most of the data is deleted because it is above the sea surface lt 0 Where the sub dives the range becomes constant and is not tracking the sea surface Also the beams are appearing very non perpendicular to the track M316 16 03 03 Continental shelf off PIB Swath up data available It s now looking up and understand that it is looking up For section at a reasonable depth most of the data is deleted because it is above the sea surface lt 0 Where the sub dives the range becomes constant and is not tracking the sea surface Also the beams are appearing very non perpendicular to the track M317 17 03 03 Continental shelf off PIB Short section of swath up available Looks good but data is being deleted where it is above the sea surface lt 0 M318 18 03 03 Continental shelf off PIB Square box of good data Looking up and with a depth sensor offset of 5 m to make sure the sea surface is accepted Still with misalignment of beams to track on one leg M319 18 03 03 Continental shelf off PIB No swath Clobbered by ship Profiler dead No useable swath data because navigation is strange M320 20
134. ould be used for to evaluate each file from a particular mission At this stage one should assess bottom echo and the likely value of the data Below is an example of my file log for mission M309 DATA0001173 jsf Possible weak returns at 48 95 145 metres multiples check sub depth DATA0001174 jsf Possible weak returns at 48 95 145 metres multiples check sub depth DATAO0001175 jsf No visible echo DATAO0001176 jsf No visible echo DATA0001177 jsf Solid and repeating bottom return DATA0001178 jsf Solid and repeating bottom return DATA0001179 jsf Short file but solid bottom return DATAO0001180 jsf Non repeating bottom return DATAO0001 181 jsf More non repeating bottom return DATAO0001182 jsf No visible bottom echo but funny straight echo in second part of file DATA0001183 jsf No visible bottom echo but funny straight echo continues DATAO0001184 jsf Couple of strange and steep returns noted one at 80 120 m and other at 20 30 m Hyperbolic Possible whale there are known to be whales in the area check depth DATAO0001185 jsf No visible echo From this brief assessment I concluded that the data from files DATA0001177 81 were appropriate for detailed analysis Using the capture facility I concatenated and printed these files It later became clear that this sequence contained a gap in the data but I wasn t aware of this from Jstar Figure 24 shows the results 99 Edgetech Subotom Profiler M309
135. panel is an enlargement of the boxed area in the upper panel 31 Figure 9 Temperature salinity and dissolved oxygen sections perpendicular left and parallel right to the shelf break 32 JR84 CTD Operations Summary This section of the report describes the method of acquisition and calibration of 44 CTD stations collected on JR84 The system performed excellently throughout the cruise with no serious problems encountered For all CTD stations the 2 dbar averages of the downcast data are reported as the final product The CTD equipment The CTD unit used for the measurement program was a Sea Bird 911 plus with dual temperature and conductivity sensors an altimeter dual SBE 43 oxygen sensors and a Chelsea instruments Fluorometer The configuration and serial numbers of the sensors used are in table 1 below A copy of the full calibration coefficients for the CTD is in the appendix to this section Table 1 CTD configuration throughout JR84 CTD sensor Serial Number date last calibrated Sea Bird 911 plus 09P 15759 0480 Series 410K 105 067241 30 Jun 2000 Digiquartz pressure transducer Primary SBE 4 042289 19 Jul 02 conductivity sensor Primary SBE 3 plus 032366 19 Jul 02 temperature sensor Primary pump SBE 5 051813 submersible pump Secondary SBE 4C 019112 19 Jul 02 conductivity sensor Secondary SBE 3 plus 032191 19 Jul 02 temperature sensor Secondary SBE 5 651807 submersible pump Tritech
136. r sensible swath data The lack of navigation data appeared to be due to a change in the LONWorks controller software that was intended to correct backward jumps in the position timestamp Reverting back to a previous version of software corrected the problem Missions 322 323 and 324 gave no usable data due to a faulty connection in the transmit transducer cable If we had been able to run further missions it might have been a good idea to set the sound speed instrument parameter to be something closer to the real sound speed The instrument assumes a sound speed of 1500ms while the real speed was closer to 1450ms The definition of the minimum and maximum depth parameters also needs to be made clearer as they seem to behave differently when the sonar head is inverted The documentation appears to have been written as a reference manual rather than as a user manual and assumes that the user is familiar with the system already Essential basic procedures are only mentioned in footnotes or at the end of the manual for example the first procedure needed to translate raw data isn t explained until page 400 of the Neptune manual and data logging is only mentioned as a footnote in the datagram descriptions A user guide to the instrument and associated software is desperately needed 77 AquaLAB Miles Pebody Autosub was equipped with a WS Envirotech AquaLab system that was to be used for collecting water samples during missions The AquaLa
137. rather than special 5 Found 4 had damaged pre amp in transducer housing Replaced 47 ohm resistor luckily seemed to have been the only damage 6 What was thought to be dodgy edge connector turned out to be something wrong with external trig not needed so cured by disconnection and signal processing board failing to boot properly on power up 7 Rigged battery back up supply to allow transport to fish while powered up there after leave connected with power on MISSION 309 04 03 03 1 Cable tray hanging loose on recovery 1 Replaced cable tray with plastic tube much better 2 TPII working well 92 MISSION 310 1 Could not talk to TP II for a while 1 Replaced 48V master PS unit 08 03 03 Basic Open 2 EM beacon switched to 1 min transmissions 2 TP II problem was broken lead in serial data connector at back of laptop Water Test 3 3 No digital acoustic comms 4 Onrecovery no nodes using 48V working MISSION 311 09 03 03 1 After PS replacement launched for mission 311 Before mission 311 Basic Open Water Test 4 but 48 volt current seen to be high recovered 1 Smelled chlorine around nose traced to SB connector what a mess sub 2 Replaced SB Burton 8 way with Impulse IE 55 7 way bulkhead 2 EM signal every 1 min 3 Spliced IE 55 tail into adapter lead for SB 3 Acoustic telemetry indicated firing of After mission 311 jack beacon abort dive weight 1 Replaced IE 55 connector 4 Very slow ascent 2
138. rd formats for gridded geographic data of which there are too many to mention Otherwise we need to write some basic code to reformat the xyz files into a format that can be read directly into other packages for example GeoTiff 109 Data cut off It appears that beam depths which are above sea level are automatically deleted in the acquisition by the EM2000 This might make sense in a system designed to measure seabed bathymetry however a system that is sold to be able to be configured in an upward looking configuration and that should therefore be capable of imaging the sea surface which can be several metres above notional sea level due to waves velocity anomalies errors in depth calculations this is nonsensical suggest that Simrad be asked to explain this issue Black box syndrome At present we suffer from a certain degree of black box syndrome both from the point of view of implementation of the system in Autosub and from the point of view of using the data We find that we do not have sufficient understanding of how the system works in terms of finding locking onto and tracking the bottom return identifying return echoes using the so called amplitude and phase detection schemes or why data are eliminated at the time of acquisition Although it might be claimed that such issues are commercial in confidence until some of these issues are resolved it is not clear that Simrad can demonstrate that the system is fit for purpose
139. re or the network node software On mission 309 the dropouts were particularly bad only 50 of the data were recorded Consequently we replaced the downward looking ADCP network interface electronic hardware and probably more importantly removed a serial test lead which may have been picking up electromagnetic interference This returned the good data rate to 75 but did not totally cure the problem This problem is still unresolved and will be investigated further back at SOC Water tracking range of the down ADCP was disappointing Typical ranges were 48 m to 90m for the downward looking ADCP The upward looking ADCP gave comparable or slightly better ranges Whereas it is possible that the lack of range was due to lack of scatterers of the required size range this apparently poor performance warrants further investigation Both the upward and downward looking ADCPs gave spurious range returns at times particularly under the sea ice missions That the returns appeared on several of the beams at the same time and also that there was often a group of contiguous returns suggests that the ADCP was detecting real targets The consequence of these spurious range returns was that the depth controller repeatedly pitched the vehicle upwards to try and avoid the supposed collision with the seabed On one occasion there were sufficient contiguous returns at low enough range to trigger the collision avoid behaviour This problem needs further invest
140. rface before diving where there would be a high probability of striking a heavy piece of ice Figure 19 shows the response of the vehicle using a 20kg steel weight hung beneath the nose and shows a gentle descent without loss of control of pitch 73 M323A 24 03 03 Dive Response using 20kg Drop Weight 60 D ENE EEE EE lane amp e 50 PropSpeed c Besse a 9 30 E i Es 20 10 5 Stern plane i 0 0 8 a 10 20 Pitch o a 0 50 0 51 0 52 0 53 0 54 0 55 0 56 0 57 0 58 0 59 0 60 ElapsedHours Figure 19 Vehicle Dive Response using the sink weight The drop weight is a catastrophic single point of failure should it fail to drop this was mitigated by holding the weight on a corrodable magnesium link with a total cross sectional area of 4mm the plan being that the link would corrode and drop the weight before the vehicle went beyond its safe working depth When placed in sea water at temperatures around 0 C the link took between 30 and 60 minutes to break with a 100N load This is rather too long and too variable to provide an effective fail safe mechanism Although the weight never failed to drop through firing the electr
141. rom down wind The second pass was a little too distant to recover the stray line but the third pass was successful The thin line could not be held initially and was made fast until the ship could drop back to relieve the tension In the meantime a lump of ice 2 thirds the length of Autosub drifted into and became entangled in the lines but luckily freed itself and the main recovery lines were attached in the normal way to the line leading aft When passing close to the counter Autosub appeared to accelerate towards the ship and disappeared briefly under the counter Recovery was completed during which the forward recovery line was partly cut when snagged by the damaged vehicle but luckily remained intact The sub sustained serious damage Damage assessment 1 Transmissometer lost fell out as sub lifted out of water 2 Primary CT sensor swinging under sub during recovery both C amp T damaged but look repairable 3 SeaPam transducer hanging under sub during recovery looks OK but need to check lead 4 Fluoremeter connecting tube broken 5 Edgetech transmitter transducer oil filled boot ripped off ingress of water may have seriously damaged ceramic element and tuning coil rinsed with fresh water oil impregnation possible saved ingress 6 Edgetech receiver array port torn from back plate but may be OK 7 Mesotech forward looking sonar hit by hard object but probably OK 8 Top panel minor damage 9 Port panel badly damaged
142. ruise Autosub Data Analysis Edgetech FS AU sub bottom profiler Assessment of Simrad EM2000 multi beam sonar Autosub under ice missions CTD water sampler and ADCP Cruise Track Plots 11 20 23 24 24 26 27 29 30 33 46 47 50 51 51 51 52 59 62 62 67 68 70 72 75 76 78 81 83 85 88 89 92 98 108 113 122 128 Introduction Adrian Jenkins PSO British Antarctic Survey Summary JR84 was the first cruise of the NERC Autosub Under Ice AUI thematic programme The aim of the programme is to investigate the marine environment of floating ice shelves in Antarctica and Greenland using Autosub the autonomous submersible vehicle developed at the Southampton Oceanography Centre The target area for the first cruise was Pine Island Bay in the eastern Amundsen Sea with the main focus being autosub deployments beneath Pine Island Glacier Cruise participants included the Autosub technical team the AUI science coordinator and scientists working on four separate projects funded by the programme Evolution and impact of Circumpolar Deep Water on the Antarctic continental shelf Principal investigator Adrian Jenkins British Antarctic Survey Autosub investigation of ice sheet boundary conditions beneath Pine Island Glacier Principal investigator David Vaughan British Antarctic Survey Marine geological processes and sediments beneath floating ice shelves in Greenland and Antarctica investigatio
143. sen Sea continental shelf break at 113 115 W The marine geological interest here was in the connection of the trough to the deep inner shelf regions near Pine Island Bay and the record of past ice stream activity From an oceanographic viewpoint the trough appears to act as a conduit by which warm Upper Circumpolar Deep Water is guided onto the shelf and possibly flows all the way to Pine Island Bay Scientific Party Personnel from British Antarctic Survey Natural Environment Research Council High Cross Madingley Road Cambridge CB3 OET Pat Cooper Electronics Dan Hayes Oceanography Adrian Jenkins Oceanography Jeremy Robst Computing David Vaughan Glaciology Doug Willis Computing Personnel from Southampton Oceanography Centre University of Southampton Waterfront Campus European Way Southampton SO14 3ZH Jon Copley AUI Science Coordinator Steve McPhail Autosub Nick Millard Autosub Miles Pebody Autosub James Perrett Autosub James Riggs Autosub Pete Stevenson Autosub Andy Webb Autosub Personnel from Department of Earth Sciences The Open University Walton Hall Milton Keynes MK7 6 Chris Banks Sea ice Mark Brandon Oceanography Sea ice Sarah Hardy Oceanography Ziggy Pozzi Walker Oceanography Personnel from Scott Polar Research Institute University of Cambridge Lensfield Road Cambridge CB2 1 ER Toby Benham Marine geology Julian Dowdeswell Marine geology Jeff Evans Marine geology Colm O Cofaigh Marine
144. spended for an Autosub test mission M319 involving dives to increasing depth to establish a safe depth limit for vehicle operations The mission was successful but during recovery in strengthening winds Autosub was knocked by the ship causing damage to both the vehicle and some of its sensors The swath survey continued for the remainder of the day and into the night 19 Mar 2003 The swath survey of the trough was completed by morning and the ship returned to the 2000 m isobath at W113 40 to run another CTD section stations 030 035 up onto the shelf The swath mapping of the upper slope was used to place two of the stations over the deep gullies that descend from the mouth of the trough With this complete the ship started on a swath survey of the upper continental slope as far west as W115 20 Mar 2003 Part way along the return leg of the swath survey the ship headed out to the 2000 m isobath to commence a final CTD section stations 036 040 back onto the shelf With this complete the ship headed south to find calm water within the pack ice for a buoyancy test on the repaired Autosub With the buoyancy checked the vehicle was sent on a successful dive to 300 m and back M320 Movement of the ice forced a recovery of the vehicle and the ship headed back to the shelf break The swath survey was completed back to the deepest part of the trough then the ship headed to deeper water to extend the swath mapping and CTD sections in this regio
145. t 84adpexec4 54 Input files 84adp cal 84bot cal Output files 84adp abs 84bot abs BOX 1 Method of derivation of the calibration coefficients A and 2 T Periods when the ADCP gave bottom track velocities i e when the ship was working in water depths generally less than 500m were identified The files with bottom track velocities were then calibrated with a nominal scaling in 84adpexec3 by setting the scaling factor A to one and the misalignment angle to zero The two minute ensembles of ADCP data were then merged with bestnav position fixes From these bestnav fixes the ship s east ad north velocity over ground were calculated Time periods within each data file were then identified where the ship s heading and velocity did not deviate greatly over a period of at least 6 minutes The ADCP bottom track velocities were then multiplied by 1 as the velocity of the ship given by the bestnav fixes is in the opposite sense to the velocity of the bottom as derived by the ADCP Values for A and for each time period were then derived using vector mathematics and the following formulas A Where Uapce is the bottom track ADCP derived ship speed and Uers is the GPS position fix derived ship speed that is ship speed over ground Gers Where is the direction of motion of derived from the GPS navigational fixes and ancr is the
146. ter mass properties in the deep are close agreement It should also be noted that neither of the data sets has been processed to account for a post cruise calibration or salinity correction using a water samples and salinometer Initially it was thought that the temperature and conductivity sensors lagged the pressure sensor and this may be the case The pressure is recorded instantaneously while the temperature and conductivity are measured after the water has been pumped from the intake to the sensors An estimate of this time delay can be made with the flow rate and pipe diameter The effect is not visible in our data Some casts seem to show a depth offset but it is the same on both the up and down casts while one would expect the offset to be of opposite sign On these missions the offset is also much larger than can be explained by such a lag Perhaps the effect would be measurable if a side by side cast were made Even after adjusting for a time delay the water entering the ports could be at a different depth than the pressure sensor particularly while diving or climbing This can be accounted for using vehicle pitch angle once the distance between ports and pressure sensor are known 122 Pressure dbar Figure 35 Pressure dbar Figure 36 Conductivity Autosub m316 and JR84ctdO1 1 300 400 500 600 700 800 900 1000 1100 1200 T Autosub C1 Autosub C2 1300 H JCR C1 l JCR C2 2 8 2 85
147. than normal This failed to cure the problem and so attention turned to the 1 55 19 way connectors and wiring harnesses We attempted to isolate the problem by successively removing parts of the network Eventually one connection was found that could be made to fail open circuit with light pressure applied to the harness moulding We replaced the harness and then replaced the bulkhead connector still the problem persisted At this stage we decided that there must be an endemic problem with possibly all the IE55 connectors and that we would not be able to cure the problem during the cruise Instead we proof tested the Autosub to an operating depth of 250 m mission 319 18 3 03 so that under sea ice missions could at least be carried out safely The problem with the pressure related open circuit failure of the IE55 connectors persists and we need to investigate further 88 Damage sustained during recovery after mission 319 Nick Millard Launch for Mission 319 was carried out at 1615 18 03 03 in 25 knt SW wind in sea state 4 5 A dive weight was used to avoid possible collision with a scattering of ice The sub surfaced at about 1732 by which time the wind had increased to 35 to 40 knts with sea arate 5 6 The jack in the box was fired and a close pass revealed that the recovery line was streamed nicely An attempt to back the ship up to recover the line was aborted because of the strong wind and the ship made a Williamson turn to approach f
148. the Autosub that had been repaired after the accident at the end of mission 319 Missions 321 324 consisted of under ice surveys with the Autosub being programmed to transit to a survey area navigate around a grid pattern of approximately 20km and return to a safe recovery position out side of the ice During the acoustic communication and tracking test phase of mission 322 it was not possible to locate or obtain status information of the Autosub Consequently the vehicle made its way to an emergency recovery position where it was safely taken back on board the ship Missions 323 and 324 completed successfully although significant currents during mission 324 resulted in the Autosub surfacing 8km East of its programmed recovery position Unfortunately during this time data from the upwards looking Em2000 swath system was not collected due to another connector problem 87 Connector problems Steve McPhail The JR84 Autosub campaign was marred by connector problems We noticed the first of these in mission 310 8 3 2003 A Burton 8 way connector assembly for the Seabird CTD had leaked seawater causing a short circuit of the 48 volt vehicle supply destroying a 48 volt switched mode power supply Investigation suggested that there is a problem with type of connector becoming loose after a pressurisation depressurisation cycle and combined with too tight a bend radius seawater ingress can occur between the sealing faces This problem needs further
149. the BAS Guildline Autosal model 8400B S N 63360 against Ocean Scientific standard seawater batch P141 One vial of OSIL standard seawater was run through the salinometer at the beginning and at the end of each crate of samples enabling a calibration offset to be derived and to check the 37 stability of the salinometer Once analysed the conductivity ratios were entered by hand into EXCEL spreadsheet before being transferred to the UNIX system and read into a pstar data file following the scheme detailed below o Samples Samples were taken for oxygen 18 analysis at a number of CTD stations Samples were taken by rinsing 200 ml medicine bottles three times before drying the top of the bottle with a tissue and then screwing down the cap Bottles were sealed by stretching parafilm around the neck Table 3 below shows the CTD stations where oxygen samples were taken The number of samples and their respective sample numbers are also given Table 3 Samples taken for analysis on cruise JR84 CTD Number Sample Station of Samples Number 004 10 D1 D10 006 5 D11 DI5 008 3 D15 D18 011 5 D19 D23 013 5 D24 D28 015 3 D29 D31 016 3 D32 D34 019 4 035 D38 021 5 D39 D43 023 3 D44 D46 032 4 D47 D50 033 4 051 054 034 4 D55 D58 Radon Samples Samples were taken at three stations for radon analysis Samples were taken by rinsing 200 ml plastic
150. the SSU Groups used were amp EK TOPAS The EA500 was kept in passive mode when the EM120 was operating 1 Water depths 500 700m EM120 Fixed Time 5000 ms 500 Fixed Time 5000 ms Time add on 80 TOPAS Calculated Time Time add on 50 Depth Auto 2 Water depths 500 1000m EM120 Fixed Time 3510 ms 500 Fixed Time 3510 ms Time add on 80 TOPAS Calculated Time Time add on 10 Depth Auto 3 Water depths Water depths 900 1030m EM120 Fixed Time 6000 7000 ms 500 Fixed Time 5000 ms Time add on 70 80 TOPAS Calculated Time Time add on 50 Depth Auto 4 Water depths 1400 ms EM120 Fixed Time 6000 7000 ms EA500 Fixed Time 5000 ms Time add on 0 TOPAS Calculated Time Time add on 0 multipulse on with 1000 ms time interval Depth Auto 5 Water depths 2600 2700 ms EM120 Fixed Time 7000 8000 ms 500 Fixed Time 5000 ms Time add on 0 TOPAS Calculated Time Time add on 0 multipulse on with 1000 ms time interval Depth Auto 22 TOPAS Colm O Cofaigh Jeffrey Evans TOPAS was run throughout the cruise apart from JD89 when were alongside at Rothera In the latter area repeated surveying of the same area of steep irregular topography during EM120 testing was felt to be unnecessary due to the rather poor TOPAS returns obtained We stopped logging TOPAS at 21 44 on JD93 General settings The following settings were used throughout the cruise Sampling rate 10 kHz tra
151. times are not known to better than 15 minutes When the vehicle is moving through stratified water or when the exact location of the sampled water is required this presents a serious problem that must be solved for the water sampler to be of practical use When the sub is in the deep uniform water mass below 700 m the CTD salinity is generally 0 03 0 05 psu lower than the samples The error introduced by averaging over 15 minutes is very small in this case 125 Mission Bag Number Number Sample Salinity psu 34 6966 CTD Salinity psu 34 7267 900 m 34 7388 34 7214 1200 m 34 7257 34 7253 800 m 34 7355 34 6393 400 m 34 6737 34 1209 100 m 34 2223 33 6639 mixed layer halocline 33 8228 33 5430 less than 15 min 33 5790 34 6847 N A 34 7210 CTD not logging 500 m 34 7526 34 7198 900 m 34 7449 34 7172 1200 m 34 7508 34 7227 800 m 34 7498 34 7050 500 34 3935 33 5853 mixed layer halocline 33 6386 34 5158 N A N A CTD not logging CTD not logging 34 7486 34 7181 less than 15 min 800 m 34 7408 34 7133 1100 m 34 7498 34 7127 1200 m 34 7545 34 7174 800 m 34 6839 34 6608 mixed layer halocline 34 1009 N A CTD not logging M317 8 33 4728 34 3633 N A 34 6438 CTD not logging 600 m M317 8 34 7072 34 7143
152. tion for temperature is applied in this script due to the temperature dependency of the velocity scaling correction A see later This correction was the value derived on JR55 i e 1 0 00152 temp Input files 84adp 84bot Output files 84adp t 84bot t Step 3 Time correction The DAS software time stamps the ADCP data This time stamp comes from the Pentium 2 which drifts at a rate approximately one second per hour To correct this to the ship s master clock the two clock times were read several times a day and the difference calculated The Julian date JDAY ADCP clock reading and calculated time differences were entered into the time correction file 84 start go which also runs 84adpexec0 0 1 and From this calculated time drift a correction was derived and applied to the ADCP data time using the Unix script 84adpexec Input files 84adp t 84bot t Output files 84adp corr 84bot corr NB 84 1 should be run 12 hours in arrears to allow for the corrected time falling outside of the 12 hour input file period which will cause the program to fall over Step 4 Correction for gyrocompass error The ADCP measures water velocity relative to the ship To calculate east and north water velocities from ADCP data information is required on the ship s heading and velocity over the ground This is partially fulfilled with input from the ship s gyrocompass described in th
153. to the EM120 500 and TOPAS We further note that a to our knowledge these problems did not occur on JR71 in February March 2002 which both COC and JEV participated in and helped operate the EM120 and TOPAS and b the EM120 and SSU software was upgraded subsequent to JR71 in October 2002 The principal problem that we noted relates to the EM120 returns in deep water In water depths generally from 2000 4000 m the individual ping display can show a centre track anomaly in which beams approx 77 134 maximum range are consistently much deeper on the order of 100 300 m than the outer beams on either 20 side and the centre beams 90 98 giving the appearance of a deep valley with a single large spike in the middle In some cases the valley base was so deep that the bottom was not actually recorded by the system Sometimes the anomaly consisted of a single deep trough with no spike of shallower depths in the middle Amplitude detection was very high where the anomalous depths were recorded and the data quality was poor with little or no phase detection Ship speed was generally 10 15 knots as the problem happened mostly when we were in transit across deep sea areas Initially we thought that this problem might be due to an incorrect sound velocity file being used in the survey We tested this by doing XBT s casts and uploading them to the survey However this produced no improvement and in some cases the anomaly became worse He
154. tune which can be difficult To this end XBT s were collected throughout the cruise T5 s were used in deep water gt 760 and T7 s were used in shelf areas water depths lt 760 m The casts were made from the starboard side of the ship due to access considerations The ship was slowed to a speed of 6 knots or less during 5 casts while T7 s were done at 15 knots Most of the T5 casts went to the full T5 depth of 1830 m Originally it was intended that a large amount of time would be spent surveying on the shelf in Pine Island Bay Hence a larger number of T7 s than T5 s were carried on board We had intended to carry out XBT casts every 6 9 hours throughout the course of the cruise However because sea ice prevented full access into Pine island Bay more time was spent surveying in deeper water than originally planned As a result we had to reduce the frequency of T5 casts so as not to run out prematurely before the end of the cruise In practice this was not a problem as the EM120 data did not need frequent updates to the sound velocity files which we judged based on the appearance of the beams arching upwards or downwards Deep water XBT s were generally carried out therefore every 1 2 days and the data immediately exported and applied to the current survey 24 XBT Stations Cast no EDF file Time date Lat
155. ubglacial AR Maximum bedforms Anvers Island Drumlins Antarctic Ce and scalloped bedrock Figure 4 Swath bathymetry of the Gerlache Strait western Antarctic Peninsula Streamlined glacial erosional and depositional forms are illustrated and are used to infer the direction of past ice flow 18 4 Ryder Bay amp inner Marguerite Bay Antarctic Peninsula Palaeo ice sheet flow Last Glacial Maximum Figure 5 Swath bathymetry of part of Ryder Bay and the inner eastern arm of Marguerite Bay Rothera is located Note the drumlinised streamlined bedrock 19 EM120 Colm O Cofaigh Jeffrey Evans We gratefully acknowledge useful discussions with Jeremy Robst and Doug Willis BAS ITS and Pat Cooper BAS ETS The EM120 was operated throughout the cruise apart from JD89 when were alongside at Rothera giving a total of 34 days data collection Angular coverage was set to Manual and beam spacing was set to Equidistant The beam angle used throughout the cruise varied according to sea conditions water depth and sea bed type but was usually between 50 65 degrees During surveys overlap between individual swath lines was achieved by means of the Helmsman s Display on the bridge which the Bridge Officer used to adjust the ship s course and maintain a reasonable 10 level of overlap Post processing of the EM120 data was carried out using the Kongsberg Simrad Neptu
156. ubsequent retreat history Quaternary Science Reviews 21 1879 1897 Stokes C R and Clark C D 1999 Geomorphological criteria for identifying Pleistocene ice streams Annals of Glaciology 28 67 74 15 1 Pine Island Bay Continental Margin nsw 112W Subglacial bedforms in trough h Iceberg scours across trough margin Palaeo ice sheet flow direction Outer Thwaites cross shelf trough Figure 2 JR84 swath bathymetric coverage between 107 and 115 W on the Amundsen Sea continental margin offshore of Pine Island Bay Gullies and chutes can bee seen on the upper continental slope The outer shelf is scoured by icebergs at shallower depths and glacial lineations are also observed in the Thwaites Trough 16 2 Thwaites cross shelf trough Iceberg scours across trough margin Palaeo ice 113 W sheet flow 111 direction 7140 S Metres Subglacial 400 500 600 700 lineations in trough 1 71 50 S Palaeo ice sheet flow direction Increase in iceberg scours as trough ae progressively shallows to the west 112 W Figure 3 Swath bathymetry showing glacial lineations in sediments of the Thwaites Trough The direction of past glacier flow can be inferred from the orientation of the lineations Icebergs scours are present in shallower areas 17 3 Gerlache Strait Palaeo ice sheet flow gt direction Streamlined e Last Glacial s
157. ulse to be synchronised with other systems on the vehicle in order to control interactions between instruments This controller also sends time and navigation information to the instrument A second LONWorks controller sends attitude and depth information to the instrument This system was initially fitted with the transmit transducer mounted in the nose of the Autosub vehicle and the receive transducer mounted in the tail section facing downwards The transducers were mounted behind polythene windows in the vehicle s fibreglass outer panels Missions 307 308 and 309 were run with this configuration The data collected showed reduced seabed depths at the outer extremities of the swath There was also a problem with poor resolution in the position information sent from the Autosub vehicle to the EM2000 This second problem was solved by a software modification to the LONWorks controller The system was then reconfigured for under ice work with the transducers looking upwards The appropriate installation settings for sensor roll were also changed SIR and S2R were set to 180 0 During the next few days a number of missions were run for testing purposes that included a short horizontal run at 100m depth in order to try to obtain data from sea surface reflections Since the sea surface should present a flat surface to the swath system it was considered to be a good diagnostic test to see if the reduced ranges at the edge of the swath were still seen
158. unications with the vehicle and it is therefore disabled whenever these communications are taking place This may have resulted in the gaps in data mentioned by David Vaughn in his section of the cruise report For the JR84 cruise the ping interval was set to two seconds The instrument only gave useful data during the first three missions Later missions flew too far above the seabed to produce any useful data The transmit transducer was seriously damaged after mission 319 and the whole instrument was disconnected subsequently in order to save power Future Improvements Currently no navigation or attitude information is stored with the FS AU data as it was understood at the time of system integration that there was no method of reading this data Edgetech have recently announced that they will be producing software that can use any navigation data stored with profiler data It would require a simple wiring addition to send navigation data to the profiler so this may be an option worth implementing for future cruises 75 Simrad EM2000 Multibeam Swath System James Perrett The Simrad EM2000 is a multibeam swath bathymetry system which operates at a frequency of 200kHz and can form up to 111 beams of data with an angular coverage of up to 60 degrees under favourable conditions On Autosub the instrument is triggered by a controller connected to the vehicle s LONWorks network This controls the ping rate and also allows the trigger p
159. utosub from open water outside the ice edge After the launch M323 another 500 m CTD station 043 was run Following the recovery of Autosub the ship proceeded eastward 25 Mar 2003 A repeat of yesterday Autosub was launched M324 in open water north of the ice edge at S70 25 W098 30 and a 500 m CTD station 044 was run The vehicle was recovered late because of navigational errors accumulated as it drifted with the current Following the recovery the ship began its transit to Rothera 26 Mat 2003 In transit Underway systems logging Slow progress in heavy seas near Peter I Island 27 Mar 2003 In transit 28 Mar 2003 In transit Cruise dinner in the evening 29 Mar 2003 Ship arrived at Rothera in the early hours and undertook some swath mapping in Ryder Bay High winds gusting to 50 knots prevented us mooring in the morning so the ship undertook further swath mapping including tests of the system under differing speed and heading Wind abating but still too strong to go alongside at Biscoe Wharf 30 Mar 2003 The ship went alongside at Rothera but the planned Autosub swath trial was cancelled because of the weather Cargo work all day 31 Mar 2003 Cargo work completed by midday local Ship departed for Jubany 1 Apr 2003 In transit Afternoon passage through Lemaire and Neumayer channels 2 Apr 2003 Cargo work at Jubany in the morning Passage to the Falkland Islands commenced after lunch 3 Apr 2003 In transit
160. wdept dist_off 84ctd001 04 03 2003 063 13 39 62 0959 64 4892 62 5 75 64 29 35 475 466 4 84 84ctd002 08 03 2003 067 14 17 102 1523 70 4582 102 9 14 70 27 49 3230 1018 999 84ctd003 10 03 2003 069 15 15 100 5158 70 4913 100 30 95 70 29 48 3525 3532 999 84ctd004 13 03 2003 072 20 13 108 9342 71 0498 108 56 05 71 2 99 2050 2061 8 79 84ctd005 14 03 2003 073 05 13 108 9368 71 1133 108 56 21 71 6 8 1502 1476 6 16 84ctd006 14 03 2003 073 07 00 108 9248 71 1482 108 55 49 71 8 89 984 986 5 28 84ctd007 14 03 2003 073 08 25 108 924 71 1814 108 55 44 71 10 89 515 525 2 78 84ctd008 14 03 2003 073 09 59 108 9284 71 2664 108 55 71 71 15 99 484 478 2 69 84ctd009 15 03 2003 074 12 55 110 3287 71 1222 110 19 72 71 7 33 2088 2066 5 93 84ctd010 15 03 2003 074 15 05 110 2314 71 1656 110 13 88 71 9 94 1544 1506 8 6 84ctd011 16 03 2003 075 05 09 113 3457 71 2231 113 20 74 71 13 39 2094 2017 4 42 84ctd012 16 03 2003 075 07 46 113 1163 71 3099 113 6 98 71 18 59 1489 1495 8 25 84ctd013 16 03 2003 075 09 34 112 9995 71 3529 112 59 97 71 21 18 1024 991 5 31 84ctd014 16 03 2003 075 11 04 112 9324 71 3762 112 55 94 71 22 57 594 578 6 37 84ctd015 16 03 2003 075 12 34 112 7708 71 441 112 46 25 71 26 46 428 422 8 45 84ctd016 17 03 2003 076 02 27 114 2756 71 732 114 16 53 71 43 92 482 468 2 76 84ctd017 17 03 2003 076 04 03 114 0445 71 7126 114 2 67 71 42 76 558 536 4 98 84ctd018 17 03 2003 076 05 29 113 7719 71 71

JR84 Cruise Report - British Oceanographic Data Centre

Contents

Download Pdf Manuals

Related Search

Related Contents