Home

Kenai River Chinook salmon sonar assessment

Contents

1. j J 2011 Behind the offshore TX data Ao aa sae Gan Come i e A JL 2011 05 14 0134 starrer 26 2011 JL 2011 05 15 0135 JL DIDSON Field Mamual in progress 22 2011 10 22 A File fate A Echoview11 Data Analysis Computer Ill a 05 17 JL 2021 05 18 10138 ara Static IP for NSD 192 168 1 105 i 2011 0519 0139 Aperl DFGANCDSF158144 JL 2011 05 20 J0140 p s J 2011 05 21 0141 J 2011 05 22 JD142 J 2011 05 23 JD143 Username FISHSONAR Password Chinook1 A Shortcut to Batch GO AL Data2011 2011 05 16_ D136 gt LBank Stratum3 Setl Organize Burn New folder 4 Z didson10 192 168 L101 public N Name Daba m dillad J 2011 14 5mile prospective site J 2011 Establish thresh JL 2011 Project OperationalfPlan J 2011 Training Files W 2011 05 16 000000_L1 4df 2011 05 16_000000_LLF ech W 2011 05 16_010000_LLF ddf 2011 05 16_010000_LLF ech Data Analysis Computer Ill E Y 2011 05 16 020000_LLE ddf Static IP for NSD 192 168 1 103 Pa E 2011 05 16_020000_LLF ech DFGANCDSF160902 a adas gr W 2011 05 16 030000_LLF ddt 2011 05 16_030000_LLF ech Y 2011 05 16_040000_LuF cat 2011 05 16 _ 040000_LLEech Y 2011 05 16 050000_LiF cat 2011 05 16_050000_LLF ech Username FISHSONAR Password Chinookt AL 2011 Behind the off 2011 05 14 J013 A 2011 05 15J0135 2011 05 16 JD136 TX data FISH Ru W 2011 05 16 060000_LL dot 2011 05 16 060000_LLF ech J Stratuml
2. 23 111 Figure 8 RM 8 6 DIDSON data storage and file management configuration for the Network Storage Device at the Soldotna Office showing the contents of the daily subdirectory for May 16 2011 For each day there are Left and Right bank subfolders each of which has subfolders for four range strata Stratum 1 3 3 8 3 m and stratum 2 8 3 13 3 m contain a single data set There are two data sets and folders for range strata three 13 3 23 3 m and four 23 3 33 3 m because these range strata are sampled twice during two separate 10 min penods See RIUS O es 36 Figure 9 RM 8 6 DIDSON data Storage and file management configuration for Network Storage System in the Soldotna office showing the contents of the fish subdirectory for May 16 2011 The Fish sub directory for each day contains the txt files containing manually measured fish lengths These files are 1 processed locally using SAS to generate daily estimates of fish passage and 2 uploaded to an Anchorage server for further processing by a Biometrician based in Anchorage or Fairbanks 6 37 Figure 10 Example of high density passage at RM 8 6 on July 20 2011 There are approximately 150 fish in this 2 5 minute chart recording A 10 minute sample with such high density can take several hours to process if each individual fish must be measured c 0ooooocnccccnoooooooooooooooonon ono nononnnnnnnn no non no nono nono nono nono non nono
3. Measure all salmon shaped fish Length M ll sal DL gt 75 Measure all salmon Measurements as n Measure a sample of fish 40 lt DL lt 75 shaped fish DL gt 75 DL P Count remaining fish 40 lt DL lt 75 Delete incidental DL lt 75 Direction of All salmon shaped fish Record DoT for all fish DL gt 75 Record DoT for all fish Travel DoT DL gt 40 Do not record DoT for fish DL lt 75 DL gt 75 Salmon y YFT Upstream Eqs 1 4 Up DL gt 75 Estimate Up DL gt 40 Up Dn 40 lt DL lt 75 TONDO IS y Biased slightly high Chinook Z y Tc Z Yer TFT Upstream Mixture model on Mixture model O oii Estimate Up DL gt 40 Up DL gt 75 Yer y y Z Unbiased Large Fish Upstream k gt es Eqs 1 4 Eqs 1 4 Eqs 1 4 mi Up DL gt 75 Up DL gt 75 Up DL gt 75 Net Apportioned m san Upstream U pa si Not possible tais Egs 5 6 Biased slightly high W Flatfish seals and beluga whales will not be recorded RM 9 DATA COLLECTION SONAR EQUIPMENT AND CONFIGURATION Components of the DIDSON system at river mile 8 6 are listed in Table 2 DIDSON operates at two frequencies a higher frequency that produces higher resolution images and a lower frequency that can detect targets at further ranges but at a reduced image resolution Two DIDSON models are currently available based on different operating frequencies For this study because we require high resolution images to at least 30m we selected the long range DIDSON model DIDSON LR that can ope
4. Funded Man PCN Name Title Months 4166 James D Miller Fishery Biologist II 12 0 4234 Michael Friedrich Fishery Biologist I 3 3 1793 Mike Hopp Fish amp Wildlife Technician III 3 3 4146 Alex Pettey Fish amp Wildlife Technician H 3 0 1045 Aaron Gordon Fish amp Wildlife Technician II 3 1 4305 Nathan Plate Fish amp Wildlife Technician I 3 0 la The total proposed personnel requirements for the new ARIS based Sonar project at RM13 7 1s are summarized in Table 9 Table 9 Project personnel for RM 13 7 site Funded Man PCN Name Title Months 1960 Debby Burwen Fishery Biologist III 12 0 N13029 Brandon Key Fishery Biologist I 9 0 B216 Kara Bethune Fishery Biologist I 2 0 5361 Lindsay Fagrelius Fish amp Wildlife Technician I 3 0 5347 Cyndarienne Jaffa Fish amp Wildlife Technician II 3 0 29 0 26 REFERENCES CITED Baumgartner L J N Reynoldson L Cameron and J Stanger 2006 Assessment of a Dual frequency Identification Sonar DIDSON for application in fish migration studies NSW Department of Primary Industries Fisheries Final Report Series No 84 Narrandera New South Wales Belcher E O W Hanot and J Burch 2002 Object identification with acoustic lenses Pages 187 192 in R Werner editor Proceedings of the 2002 International symposium on Underwater Technology Institute of Electrical and Electronic Engineers Piscataway New Jersey Bromaghin J 2005 A versatile net selectivity model with application
5. SAMPLING PROCEDURES Similar to the RM 8 6 site a systematic sample design Cochran 1977 will be used to sequentially sample discrete range strata range windows for a total minimum of 10 minutes each Dividing the total range to be insonified into shorter range strata allows the aim of the sonar beam to be optimized for sampling a given river section i e generally the aim must be raised in the vertical dimension as sections further from shore are sampled 14 A fixed sampling schedule similar to that shown in Figure for the RM 8 6 site will be developed for the RM 13 7 site at the start of the season after determining how many discrete range strata will need to be sampled The number and length of range strata required to optimally sample the river cross section at RM 13 7 have not been determined ARIS is not limited to fixed range windows like DIDSON approximately 1 25m 2 5m 5m 10m 20m so there is more flexibility in setting a range stratum length ARIS also allows the user to set the number of samples per beam to a maximum of 4 000 samples whereas DIDSON is limited to 512 samples meaning an ARIS 40m range window collected at 4 000 samples provides the same downrange resolution as a DIDSON 5m range window At close range a short range window of 5m will likely be used for better focal resolution but ranges beyond about 10m can be collected at any desired length as long as the aim 1s appropriate over the entire length o
6. The resolution of a DIDSON or ARIS image is defined in terms of down range and cross range resolution where cross range resolution refers to the width and down range resolution refers to the height of the individual pixels that make up the image Figure A1 2 Each image pixel in a DIDSON or ARIS frame has x y rectangular coordinates that are mapped back to a beam and sample number defined by polar coordinates The pixel height defines the down range resolution and the pixel width defines the cross range resolution of the image shows that image pixels are sometimes broken down into smaller screen pixels e g pixels immediately to the right of the enlarged pixels an artifact of conversions between rectangular and polar coordinates Cross range resolution The cross range resolution is primarily determined by the individual beam spacing and beam width both of which are approximately 0 3 for the DIDSON LR HRL at 1 2 MHz Table Targets at closer range are better resolved because the individual beam widths and corresponding image pixels increase with range following the formula below X 2Rtan 6 2 1 Where X width of the individual beam or image pixel in meters R range of interest in meters and 0 individual beam angle in degrees approximately 0 3 Down range resolution Window Length 1 e the range interval sampled by the sonar controls the down range resolution of the image and is calculated using
7. number of salmon shaped fish gt 40 cm DL except for downstream fish gt 75 cm DL in stratum s for hour j and day i The DL mixture model modified to accommodate censored lengths from small fish Appendix D3 will be fit to the same subset of the data described above The daily proportion 777 of these fish that are upstream Chinook salmon output from DL mixture model will be multiplied by the estimate of yrr above to estimate z number of Chinook salmon migrating upstream in the midriver corridor Equations 5 and 6 Because direction of travel information is not available for fish 40 lt DL lt 75 it will not be possible to estimate y daily upstream passage of DL gt 40 cm salmon shaped fish Quantity yrr can be substituted to produce the net apportioned estimate however it will be biased high by an unknown small amount because it includes some downstream fish 40 lt DL lt 75 MODIFICATIONS UNDER LFO PROTOCOL Under the Large Fish Only protocol only those fish greater than 75 cm DL will be counted and measured This constraint will require the following modifications to the abundance estimators Because small fish are not counted it will not be possible to directly estimate y daily passage of upstream DL gt 40 cm salmon shaped fish nor the net apportioned estimate of which y 1s a component Daily upstream passage x of salmon shaped fish gt 75 cm DL will be estimated as specified in equations 1 4 The d
8. V Number of Beams 48 Beam Spacing 0 6 Extended Window Start 0 83m to 52 3m in 0 83m steps Extended Window Length 5m 10m 20m 40m Range Bin Size relative to window length 10mm 20mm 40mm 80mm Pulse Length relative to window length 18us 36us 72us 144us Identification Mode Operating Frequency 1 8 MHz Beamwidth two way 0 3 H by 14 V Number of Beams 96 Beam Spacing 0 3 Extended Window Start 0 42m to 26 1m in 0 42m steps Extended Window Length 1 25m 2 5m 5m 10m Range Bin Size relative to window length 2 5mm 5mm 10mm 20mm Pulse Length relative to window length 4 5us 9us 18us 36us Both Modes Max Frame Rate range dependent 4 21 frames s Field of view 29 Remote Focus 1m to Infinity Control amp Data Interface UDP Ethernet Aux Display NTSC Video Max cable length 100 10BaseT 61m 152m 200ft 500ft Max cable length twisted pair Patton Extender 1220m 4000ft Max cable length fiber optics kilometers Power Consumption 25 Watts typical Weight in Air 7 9 kg 17 4 Ib Weight in Sea Water 1 0 kg 2 2 Ib Dimensions 31 0cm x 20 6cm x 17 1cm Topside PC Requirements Windows XP Vista 7 Ethernet Optional NTSC video monitor DIDSON LR Specifications Detection Mode Operating Frequency 0 7 MHz Beamwidth two way 0 8 H by 14 V Number of Beams 48 Beam Spacing 0 6 Extended Range Settings Extended Window Start 0 83m to 52 3m in 0 83m steps Extended Window Length
9. W 2011 05 16 070000_LUF at 2011 05 16_070000_LLF ech Y 2011 05 16_ 080000_LUF adit J Stratum a Stratum3 Data Analysis Computer Ill A seu A so 2011 05 16_080000_LLF ech pa Y 2011 05 16 090000_LL5 ddf Static IP for NSD 192 168 1 104 a neh 2011 05 16 090000_LLF ech DFGANCDSF 160898 Aalst W 2011 05 16 11000 1 Username FISHSONAR Mas wane 16_110000_LHFech 2011 05 16_120000_LHf ddf Password Chinookl E 2011 05 16_120000_LHF ech E Shot W 211 05 16_130000_LHf ddf 2011 05 16_130000_LHF ech 2011 05 17 0137 W 2011 05 16_140000_LHF cdf J 2011 05 18 0138 aa e mon 46 items Offline status Online Offline availability Not available Mi Figure 8 RM 8 6 DIDSON data storage and file management configuration for the Network Storage Device at the Soldotna Office showing the contents of the daily subdirectory for May 16 2011 For each day there are Left and Right bank subfolders each of which has subfolders for four range strata Stratum 1 3 3 8 3 m and stratum 2 8 3 13 3 m contain a single data set There are two data sets and folders for range strata three 13 3 23 3 m and four 23 3 33 3 m because these range strata are sampled twice during two separate 10 min periods see Figure 5 36 Network Storage System Western Digital ShareSpace 8TB IP 192 168 1 102 DFGANCDSFDSSHR2 Mapped to M drive Access using Username admin Password Chinook 1 GB Network Switch Netg
10. River Mile 13 7 KPB Parcel No 055 250 30 Sheet 10 of 11 Date 3 Apr 13 Removable elevated light penetrating walkway 3 ft wide and steps 3 ft x 11 ft Ordinary High Water N Figure F1 12 Side view of minor channel battery bank and walkway deployment at the RM 13 7 sonar site 108 Side View Minor Channel Left Bank looking downstream Weir and Tripod Deployment 45 ft rigid tripod style Submerged steel tripods picket weir to deflect fish son from OHW with to pass in front of sonar Large warning units attached location oe eer buoys suspended will be identified by a from weir tripod marker float on the water s supports surface TW ES TTT AN TIA Figure F1 13 Side view of minor channel battery weir and sonar tripod deployment at the RM 13 7 sonar site 109 RM 13 7 Tent Electronics Schematics amps max i A l k Milti Plug Apapter J AC power from cabin 0 45 amp 4 Lo Pr mas a md S Figure F1 14 Estimated power requirements for electronics based in the main camp on left bank 110 APPENDIX G IP ADDRESSES 111 Appendix G1 RM 8 6 sonar site IP addresses located 17 purpose SONARSITEatRM86 All subnet masks 255 255 255 0 RM8 6 J Synapsis NetBooter Left bank 128 95 97 150 RM 8 6 Remotely on off Left on off switch bank DIDSON O AO D D V O E Local Tactical SubNet MA Tactical Computer one 192 168 1 2 RM 8 6 SonarUse
11. t2 of 11 Dai Two 300 Data Cables 8 x 8 platform w battery bank s Two ARIS Topsides Two Wireless Hubs and Remote Power Switch Tote w battery bank ARIS Topside Wireless Hub and Remote Power Switch N 300 Data Cable srs 30 ft 98 PS 5 250 23 055 250 30 ADF amp G Sport Fish A Sonar Site Investigation Study River Mile 13 7 Fa KPB Parcel No 055 250 23 Sheet 3 of 11 Date 3 Apr 13 Ordinary na Water High NORTH Removable elevated light penetrating walkway 3 ft wide platform 12 ft x 8 ft and steps 3 ft x 8 ft River Flow Submerged steel tripods 76 ft and 24 ft from OHW with ARIS units attached location will be identified bya marker float on the water s surface Figure F1 3 Aerial view of main channel left bank camp and sonar deployment at the RM 13 7 sonar site 99 Side View Main Channel Left Bank looking downstream Tent Walkway Platform and Steps 10 ft x 12 ft WeatherPort supported by elevated temporary wooden plywood platform located in open area so no vegetation removal will be required Tent located 50 ft inland from ordinary high water Removable 12 ft x 8 ft removable ELP elevated light platform penetratingsteps Removable ELP walkway 3 ft wide 3 ft x 11 ft Figure F1 4 Side view of main channel left bank camp and walkway deployment at the RM 13 7 sonar site 100 Side V
12. 00 80 5058 FYSVOSRRIU SR LANAS TITE HERREN ELLE lle leete Ek meters Fish 1 Range 164m Length 115 4 or Midness 0 0 Erker gt gr usa Y bo add list Tam ce vs ta Figure C1 2 DIDSON images from a tethered Chinook salmon showing the original DIDSON image top the zoomed image middle and the segmented lines that result when the observer clicks along the length of the fish to mark its length bottom Adapted from Burwen et al 2010 71 APPENDIX D DIDSON LENGTH MIXTURE MODEL AND ASSOCIATED BUGS PROGRAM CODE 12 Appendix D1 Mixture model for estimating species composition of migrating fish Mixture models are useful for extracting information from the observed frequency distribution of a carefully selected measurement If one were able to observe the exact length but not the species of every fish passing the sonar in the Kenai River the distribution of such measurements might look something like Figure D1 1 Given additional knowledge about the size of sockeye and Chinook salmon the shape of the overall distribution can reveal much about the relative abundance of sockeye and Chinook For instance if it is known that sockeye salmon do not exceed 70 cm in length and that small Chinook are very rare one can conclude that the left hand mode of the distribution 1s composed almost entirely of sockeye salmon and that the species composition 15 perhaps 50 50 sockeye Chinook Mixture model analysis i
13. 1 lt 25 Lsig 2 2 lt 25 Lsig 2 3 lt 25 for s in 1 2 for a in 1 3 Ltau s a lt 1 Lsig s a Lsig s a mu 1 1 dnorm 621 0 0076 mu 1 2 dnorm 825 0 0021 mu 1 3 dnorm 1020 0 0047 mu 2 1 dnorm 380 0 0004 mu 2 2 dnorm 500 0 0004 mu 2 3 dnorm 580 0 0004 for ain 1 3 pa effective 1 a lt pa 1 a q1 a a inprod pa 1 q1 al pa effective 2 a lt pa 2 a for k in 1 5 TL cm 75 k lt TL cm k 75 mu DL1 k lt beta0 beta1 TL cm 75 k DL1 k dnorm mu DL1 k tau DL for i in 1 n_fish age i dcat pa effective species i 1 3 mefl mmi i dnorm mu speciesf i age i _tau species i agef i continued 80 Appendix D3 Page 2 of 2 for j in 1 n_meas species2 j dcat ps age2 j dcat pa species2T j 1 3 mefl mm 2 j dnorm mu species2 j age2 j Ltau species2 j age2 j TL2 cm 75 j lt 1 1 mefl mm 2 j 2 10 75 CONVERSION TO TL BASED ON NUSHAGAK 2001 DATA mu DL2 j lt beta0 beta1 TL2 cm 75 j DL2 j dnorm mu DL2 j tau DL for k in 1 n_small species3 k dcat ps age3 k dcat pa species3 k 1 3 mefl mm 3 k dnorm mu species3 k age3 k _tau species3 k age3 k TL3 cm 75 k lt 1 1 mefl mm 3 k 2 10 75 mu DL3 k lt beta0 beta1 TL3 cm 75 k DL3 k dnorm mu DL3 k tau DL I 70 81 Appendix D4 OpenBUGS code for daily abundance model to be used when it is not possi
14. 130 Fish Length cm Figure D1 1 Hypothetical distributions of fish length measurements black solid lines from the Kenai River a top Few small Chinook salmon no measurement error b bottom 40 of Chinook salmon are small 75 measurement error standard deviation 10 cm Distributions of sockeye red dashed line and Chinook blue dashed line true length are shown in case b In both cases the true species composition is 50 sockeye 50 Chinook DIDSON VS Actual Length Observer 1 y 0 8364x 9 8661 R2 0 9273 DIDSON Length cm Observer 2 y 0 8467x 7 3544 R2 0 9157 50 60 70 80 90 100 110 120 Measured Fork Length cm Figure D1 2 Relationship between DIDSON based length measures and fork length for two independent observers in 2007 Results are from 37 tethered salmon insonified by a DIDSON LR sonar with a high resolution lens at RM 8 6 site 76 f Species Proportions a Sockeye b Sockeye cm Length cm Frequency Frequency Data oe Model 120 20 40 0 80 100 120 J g DIDSON Length cm 120 7 c DIDSON length 100 vs Length gt 2 D c o ba Z O J a O 20 40 60 80 100 120 Length cm 0 20 40 60 80 100 120 d Chinook e Chinook DIDSON Length cm Frequency 40 60 80 100 120 20 40 60 80 100 120 Figure D1 3 Flow chart of the DIDSON length mixture model described in the text The frequency distribution of DIDSON length mea
15. 16_003000_RHF ddf Laptop Hard Drives a jE Computer a E Os C Le dell Y 2012 05 16_004000_RHF ddf 2012 05 16_005000_RHF ddf Y 2012 05 16 010000_RHF ddf a Je Didson Data J WinHelp zip All DIDSON data are stored in the Didson Data subdirectory Figure 7 DIDSON data storage configuration and directory structure at RM 8 6 site DIDSON sample files are initially stored on the data collection computer then transferred to the external hard drive using a batch file that also appends an L or R to each filename to indicate which bank the file originated on 35 Network Attached Storage Synology DiskStation 1512 8 TB IP 192 168 1 165 DFGANCDSFDSSHRUpdate Mapped to M drive 1 GB Network Switch Netgear ProSafe 8 Port Gigabit Switch model GS108 ae Access using Username Admin Password Chinook 1 GB ethernet Cat 6 cables G Q F Computer didson10 192 168 1 101 public N we 4 Search didson10 111921681 P Organize 3 Open Burn New folder iv O 4 y Gidsonl0 11192 168 1101 public N Name i Date modified Type Dell E6500 Ik 2011 145 mite prospective site e J 2011 145mile prospective site J 2011 Establish thresh J 2011 Establish thresh J 2011 Project Operational Plan a p O ps M 2011 Project Operational Plan 2011 Training Files J 2011 Training Files os J ARIS 2011 or whatever EL Baich pes J Copy to Desktop ese i 4h Data2011 R e
16. 51 45 07 52 00 07 52 15 Current 237 Last 9942 Figure 10 Example of high density passage at RM 8 6 on July 20 2011 There are approximately 150 fish in this 2 5 minute chart recording A 10 minute sample with such high density can take several hours to process if each individual fish must be measured 38 E gt A Y busting through 2012 08 14 210000_LHF ddf DIDSON Control and Display V5 25 l Sy lt File Edit View Image Sonar Processing Aux Help Yee ever dre rigs ode ear 0 g El VRE WORD ESE EE SES AVY w a Sonar Controls Frame Rate fe Total Frames 101 Receiver Gain 40 E Window Start 3 33 H Window Length 5 00 T Focus 5 87 M AutoFreq LF M AutoRate HF Display Controls M Reverse Grid M Smooth 7 Measure Palette LCD2 v Intensity File Position Frame 56 00 00 07 meters For Help press FL _ DEMO MODE STILL LOOP Totak101 First 0 Current 56 Last 100 Figure 11 Example of 110 cm Chinook salmon swimming past milling pink salmon at RM 8 6 site on August 14 2012 39 65 Google Earth a ae Eile Edit View Jools Add Help Figure 12 Location of nine transects conducted at the river mile 13 7 site on July 9 2012 Yellow arrows indicate preferred locations for sonars on each bank 40 Length 758 Azau 915 X 6042233 Azmutu 2698 X 6042389 Y 67105638 Y 67105048 Templa
17. 6 260 salmon measured at a nearby tagging station and found that the DIDSON based mean was 1 6 cm larger Baumgartner et al 2006 found a strong linear relationship between manual DIDSON based lengths and true lengths for golden perch silver perch and goldfish that were free swimming in a tank Burwen et al 2007 concluded that at close ranges approximately lt 12m reasonably good estimates of fish length could be obtained by manually measuring images of fish produced by a standard DIDSON operating in high frequency 1 8 MHz mode In these studies fish were insonified at relative close ranges less than 15 m with a standard DIDSON model However to adequately cover the ranges at which Chinook salmon travel at ADF amp G s established sonar site on the Kenai River high resolution images at ranges up to 30m are required Studies conducted by ADF amp G in 2007 assessed the potential for a long range DIDSON model DIDSON LR fitted with a high resolution large lens DIDSON LR HRL to provide improved image resolution at required ranges The large lens provides almost twice the resolution of the standard lens These studies concluded that sufficiently accurate estimates of fish length were now possible at ranges up to 22m Burwen et al 2010 Additional studies conducted in 2008 2009 with the same hardware but newer firmware and software extended that maximum range to 33m Miller et al 2012 With the question of obtaining fish length mea
18. Gateway OA AI 128 95 97 227 DIDSON IP Unit MAC Address XX XX XX XX XX XX Radio A MAC Address KX XX XX XK KX XX hernet or Coax to Bridge Le y i 4 ff a 7 u vw gt LL A NS vir dk i us NE n The Magnum Pure Sine Inverter kit MM51012 61000 W 12V combines the inverter and charger which are stored mounted inthe white container at the lower left corner of the fish tote The controller ME RC50 is mounted on the top of the container The system is powered by two AGM L16 6V batteries that are charged daily by 1000W Honda generator Additional IP address 128 95 97 00 batteries can be added as required Subnet Mask 255 255 010 on Left bank Figure 18 Diagram of components required on RM 13 7 right bank for wireless transmission of ARIS data back to the main camp on left bank A wireless bridge transmits data to a data collection computer on left bank for storage and subsequent transport to the Soldotna office 46 0 5 amu ARIS Topside LB Battery Nearshore Multi Backup 1 aE 16 5 amps max Plug Adapter Battery Backup 2 Battery Backup 3 Figure 19 Schematic of power requirements for RM 13 7 left bank main camp electronics 47 www gotomypc com login with Username debby burwen alaska gov password sonarianl GotoMyPC access codes for data collection host computers named by position in river plus last two digits of their computer names 2 TB External Ha
19. Medium size fish unintentionally chosen for measurement that turn out to be less than 75 cm will be counted and length will not be recorded Direction of travel will be recorded only for salmon shaped fish greater than 75 cm DL not for fish 40 cm lt DL lt 75 cm Large Fish Only LFO Sampling Protocol Occasionally milling or holding behavior can make it difficult to reliably assess the direction of travel for many fish This 1s more likely to happen during even numbered years when pink salmon are most abundant In 2010 and 2012 staff were able to easily discern larger Chinook salmon swimming directly through the smaller holding or milling salmon Figure 11 Under these conditions fish larger than 75 cm will be measured and shorter fish will be ignored The Large Fish Only LFO sampling protocol is as follows Length will be measured for all salmon shaped fish greater than 75 cm DL Flatfish seals and beluga whales will not be recorded and fish measured less than 40 cm will be omitted from further calculations Direction of travel will also be recorded for measured fish Remaining fish will not be recorded in any way due to the difficulty in ascertaining direction of travel out to be less than 75 cm will not be recorded Table 1 Sampling protocols to be applied in 2013 Medium size fish unintentionally chosen for measurement that turn Sampling Protocol Standard STD Fast Track FT Large Fish Only LFO
20. Operational Plan 2011 Training Files Dsamples2011 xt FC_2011 05 16_000000_LLF bat FC_2011 05 16_000000_RHF txt FC_2011 05 16_002000_LHF txt Data Analysis Computer Ill Static IP for NSD 192 168 1 103 See FC_2011 05 16_003000_LHF txt DFGANCDSF 160902 x pen a FC_2011 05 16_010000_LLF va Username FISHSONAR 4 dati FC_2011 05 16 010000_RHF txt FC_2011 05 16_012000_LHF txt FC_2011 05 16_013000_LHF txt FC_2011 05 16_020000_L1F ba FC_2011 05 16_020000_RHF txt FC_2011 05 16_022000_LHF txt Password Chinook1 JL 2011 Behind the offshore TX data J 2011 05 14D134 J 2011 05 15_ 0135 a 2011 05 16D136 aL FISH n DAT FC_2011 05 16_023000_LHF txt AL Didson 136 DAT FC_2011 05 16_030000_LLF tt FC_2011 05 16_030000_RHF txt A Jl LBank 7 a gt Data Analysis Computer Ill pres FC_2011 05 16 032000 LHF txt h Short FC_2011 05 16_033000_LHF txt e FC_2011 05 16_040000_LLF ba Static IP for NSD 192 168 1 104 JL 2011 05 17 J0137 AAA 2011 05 18_ D138 A g SEEE DFGANCDSF160898 lt AS EC_2011 05 16 042000 LHF bt Username FISHSONAR pto FC_2011 05 16_043000_LHF tet J 2011 05 20_ 0140 J 2011 05 21 0141 J 2011 05 22_ 0142 J 2011 05 23 0143 J 2011 05 24 _JD144 J 2011 05 25_JD145 J 2011 05 26 JD146 Password Chinook1 FC_2011 05 16_050000_LLF ba FC_2011 05 16_050000_RHF txt FC_2011 05 16_052000_LHF txt FC_2011 05 16_053000_LHF txt FC_2011 05 16_060000_LLF ba FC_2011 05 16_060000_RHF txt Text Document IKB N v 4 ni p
21. an ARIS 1200 with a telephoto lens Frames were taken from file SN1064 Kenail3 75 LB 2012 07 18 091000 T15 B48 S2000 F12 R4 22 aris approximately frames 5175 5204 The white lines following the fish mid section show the path taken on the image for the length measurements 94 APPENDIX F RIVER MILE 13 7 SITE INVESTIGATION DOCUMENTION 95 Appendix F 1 Additional details regarding river mile 13 7 sonar site Benefits of the proposed RM 13 7 site include 1 the State owns land on both sides of the river DNR land on the left KPB parcel 055 250 23 and right bank KPB parcel 057 250 30 2 the location is above major tidal influence and 3 the river channel is a relatively narrow at this location Upland on both banks is composed of trees and shrubs The main tent housing sonar electronics will be located on the left bank Figure Fl 1 Data from the right bank will be transmitted to the tent on the left bank using wireless technology Providing full bank to bank coverage will require a total of five sonars four sonars on the main channel and a fifth sonar to monitor the minor channel on right bank Figure F1 2 Proposed site diagrams for the left bank can be found in 5 Proposed site diagrams for the right bank main river channel can be 9 Proposed site diagrams for the minor river channel can be found in A 10 x12 tent will be assembled on the left bank at least 50 ft from ordinary high water levels OHW Figure F1 3 T
22. and tail Watching the behavior of the head and especially the tail over several frames and taking several measurements is often helpful in distinguishing the best frame 91 E ARIS by Sound Metrics See What Others Can t gt ES eR 0 0 dB Rectangular Palette Deep Blue Flip UR Effects None Grid Measure Geometry 7 Real time Rate Y 100 Me lt vo Count Fish n Count Fish AS i Finish Line CSOT Page P n PRE 1s ROA More Fish Data S Av 100n Fish 32 Range 8 44 Fame 5174 Bearing 990 ee 25 amp E Length 0 0 Thickness 0 0 More Del Undo Fish rs ge f lt 05 amp A 1 i gt 1 1 1 1 Q E a a MN D 1 10 09 18 20 09 18 30 09 18 40 09 18 50 09 19 00 09 19 10 09 19 20 09 19 30 Frame 5181 09 19 05 11 Up 0 Down 07 11 Total 7 Measured Frame Tracking ON eo 50m Figure E1 1 Manual measurements in ARIS echogram mode A left click on a fish trace places a red marker on the fish trace and automatically activates the movie showing the fish bounded by range arcs A vertical line defines the selected frame on the echogram and a horizontal line at the selected range shows the boundaries of the movie loop 92 TE VOR CBR ESOSEVVYESSE meters For Heb press FL DOME STL WI TEE FRO ere A11 LR Dad al x 2007 07 10_131000_1 _Sm range window dd trol d Ela G y pop ire bumn bo ih SHER 9K 4 rr ne o4d gt O18 DIA YE VORR LO ARMINIA REKEN tk
23. for implementing and maintaining wired and wireless networks used to transfer sonar data Develop batch files and other user interface programs as needed to transfer summarize and archive sonar data Assist with in season data processing and analysis Work dates May 13 June 30 2013 Steve Fleischman Fisheries Scientist I Alaska Department of Fish and Game Provide guidance on sample design and estimation procedures Assist with in season and post season data analysis Review project operational plan Shared responsibility for post season report Jiaqi Huang Biometrician II Alaska Department of Fish and Game Provide guidance on sample design and estimation procedures Assist with post season data analysis Review project operational plan and report Bill Hanot Senior Engineer Sound Metrics Provide training in the operation of a new DIDSON prototype ARIS and ensure that all hardware and software features are working correctly Develop software features required to collect or analyze DIDSON ARIS data Planned visits are May 13 17 and July 15 19 2013 Brandon Key Fishery Biologist I Alaska Department of Fish and Game 23 Duties Personnel Duties Personnel Duties Personnel Duties Personnel Duties Personnel Duties Personnel Assist Project Leader with all aspects of DIDSON ARIS deployment operation and data analysis Take lead role in developing protocol for processing DIDSON
24. nono nnnnnnos 38 Figure 11 Example of 110 cm Chinook salmon swimming past milling pink salmon at RM 8 6 site on August 42012 sis ds sd std 39 Figure 12 Location of nine transects conducted at the river mile 13 7 site on July 9 2012 Yellow arrows indicate preferred locations for sonars on each bank as 40 Figure 13 Corresponding profiles for nine transects conducted near river mile 13 7 of the Kenai Rivet 6 41 Figure 14 Aerial view of sonar sites at Kenai river mile 8 6 top and 13 7 bottom with proposed sonar beam coverage Diagrams are approximate and not drawn to scale 42 Figure 15 Proposed coverage for the right bank top transect 5 in Figure 16 and left bank bottom transect 6 m Fiume Aora RMI er Nr eR ee ene ee eee A 43 Figure 16 Aerial view of sonar sites at Kenai river mile 8 6 top and 13 7 bottom with proposed sonar beam coverage Diagrams are approximate and not drawn to scale ooooocococcncnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnos 44 Figure 17 ARIS data collection schematic for the RM 13 7 site For simplicity this diagram shows only one of three right bank data collection computer sonar pairs and one of two left bank data collection computer sonar pairs The wireless router will accommodate multiple computers cccccccnnnnnnononononcnnnnnss 45 Figure 18 Diagram of components required on RM 13 7 right bank for wireless transmission
25. oocccccccnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnininos 96 Appendix G1 lt RMES 6 sonar site IP addresses 2 as ss nes sd 112 Appendix G2 Soldotna sonar office IP dE de 112 Appendix G3 RM 13 7 sonar site IP addresses 2013 Configuration oooccccccnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnss 114 vi PURPOSE Alaska Department of Fish and Game ADF amp G has monitored Chinook salmon passage in the Kenai River using side looking sonar technology from 1987 to present The original acoustic technology dual beam sonar 1987 1994 was replaced by split beam sonar 1995 2011 and most recently by DIDSON 2011 present Starting in 2013 ADF amp G will operate two sonar stations one at the historical site at Kenai River mile 8 6 and a second experimental site at Kenai River mile 13 7 BACKGROUND Fixed location side looking sonar techniques are commonly used to obtain in season estimates of run strength for anadromous fish stocks in rivers that are too wide for installing weir structures or too occluded for visual observations Daum and Osborne 1998 Enzenhofer et al 1998 Gaudet et al 1990 Maxwell and Gove 2007 In Alaska sonar estimates of inriver passage often provide the basis for estimating spawning escapement and for regulating harvests of commercially important salmon stocks Westerman and Willette 2006 Miller et al 2010 Acoustic assessment sites currently exist on at least ten rivers in Alaska One of the
26. ppt 20 General Alaska Administrative Code all commonly accepted abbreviations all commonly accepted professional titles at compass directions east north south west copyright corporate suffixes Company Corporation Incorporated Limited District of Columbia et alii and others et cetera and so forth exempli gratia for example Federal Information Code id est that is latitude or longitude monetary symbols U S months tables and figures first three letters registered trademark trademark United States adjective United States of America noun USC U S state AAC e g Mr Mrs AM PM etc e g Dr Ph D R N etc O e g FIC l e lat or long g Jan Dec TM US USA United States Code use two letter abbreviations e g AK WA Mathematics statistics all standard mathematical signs symbols and abbreviations alternate hypothesis base of natural logarithm catch per unit effort coefficient of variation common test statistics confidence interval correlation coefficient multiple correlation coefficient simple covariance degree angular degrees of freedom expected value greater than greater than or equal to harvest per unit effort less than less than or equal to logarithm natural logarithm base 10 logarithm specify base minute angular not significant null hypothesis percent probability p
27. range stratum File size also increases with the down range resolution level Again consultant Bill Hanot will assist with determining the optimal resolution value for individual range strata We estimate that 10 minute file sizes will vary from a maximum of approximately 300 000 KB to approximately 500 000 KB Table 6 gives a worst case scenario for data storage needs at these sample rates Data from each sample will be stored to a uniquely named file Filenames are automatically generated by the ARIS software using optional identifiers such as sonar serial location bank year month day military time hour min sec transmitted pulse length number of beams sampled samples beam resolution and range interval For example the file SN 1064 Kenail3 75 LB 2012 07 17 004000 T24 B48 S2000 F12 R21 35 aris refers to a file collected by sonar 1064 at RM 13 75 on the left bank of the Kenai River that started data collection on July 12 2012 at 0 40 AM using a transmitted pulse length of 24 msec using all 48 beams with 2000 samples beam at resolution 12 over the range interval 21 35m Longer file names may be used initially but pared back 1f we determine that the long filenames hinder file handling during file transfer processing archiving But initially the additional information in the filename may help during the early phase of this project when different settings are being evaluated One laptop will be dedicated to collecting dat
28. schematic for the supplying DC power to the minor channel ARIS system at RM 13 7 via a battery bank charged by a 2000W generator 106 Top View Minor Channel Left Bank Platform Walkway Weir and Tripod Deployment ADF amp G Sport Fish E F Sonar Site Investigation Study River Mile 13 7 S Sheet9 of 11 Date 3 Apr 13 Removableelevated ae E light penetrating lt Q M gt eS walkway and steps var 9 ve NORTH Large warning buoys suspended 4 from weir tripod 45 ft rigid tripod style AN dy weirto deflect fish in front Al AM Ai a fii UT hy x MTA i mi ti hem Electronic control cable High leadingfrom topside tote Submerged steel tripod Water to sonar unitin the river 40 ft from OHW with ARIS unit attached Figure F1 11 Aerial view of minor channel battery bank weir and sonar deployment at the RM 13 7 sonar site 107 Side View Minor Channel Left Bank looking downstream Walkway and Platform Wireless radio antennae strapped to tree near west side of island to transmit data to tent located across the river up to 100 ft of Ethernet cable running from tote to wireless radio Small 1000 watt generator housed in a 3 ft x 2 ft wooden box lined with plastic for spill containment Battery bank and sonar topside equipment housed ina 4 ft x 3 ft plastic containertote T S ADF amp G Sport Fish Sonar Site Investigation Study
29. systems There are potential issues with the dual network setup that is planned in the current ARIS software Currently the ARIS software doesn t distinguish between network interfaces so if the wireless connection is enabled Files from individual range strata will vary in size if the ping rate is optimized for each range strata Since the return time for a ping increases with range off shore range strata will require a slower ping rate and files will contain fewer frames 19 the software may not connect to the sonar The workaround could be to disable the wireless interface while getting the sonars powered up and connected and then re enabling the wireless interface Note from Bill Hanot at SMC related to remote access ARIS is not configured with a static IP address It either gets an IP address from a DHCP server if present or gives itself a link local IP address If the PCs are configured as Automatic IP they will also give themselves a link local IP address Your diagram shows the wireless bridges configures with a static IP If they don t have a link local option they can be configured with a static IP on the link local subnet 169 254 xxx xxx I ll discuss this with people here If you have the manual or at least the make and model of your wireless bridge equipment that would be a help FISH LENGTH AND DIRECTION OF TRAVEL Estimates of total length will be made from images using the ARISFish V1 5 Detailed instructio
30. to Pacific salmon and freshwater species of the Yukon River Alaska Fisheries Research 74 1 3 157 168 Burwen D L and S J Fleischman 1998 Evaluation of side aspect target strength and pulse width as potential hydroacoustic discriminators of fish species in rivers Canadian Journal of Fisheries and Aquatic Sciences 55 2492 2502 Burwen D L J Hasbrouck and D E Bosch 2000 Investigations of alternate sites for Chinook salmon sonar on the Kenai River Alaska Department of Fish and Game Fishery Data Series No 00 43 Anchorage http www sf adfg state ak us FedAidPDFs fds00 43 pdf Burwen D L D E Bosch and S J Fleischman 1998 Evaluation of hydroacoustic assessment techniques for Chinook salmon on the Kenai River 1995 Alaska Department of Fish and Game Fishery Data Series No 98 3 Anchorage http www sf adfg state ak us FedAidPDFs fds98 03 pdf Burwen D L S J Fleischman J D Miller and M E Jensen 2003 Time based signal characteristics as predictors of fish size and species for a side looking hydroacoustic application in a river ICES Journal of Marine Science 60 662 668 Burwen D L S J Fleischman and J D Miller 2007 Evaluation of a dual frequency imaging sonar for estimating fish size in the Kenai River Alaska Department of Fish and Game Fishery Data Series No 07 44 Anchorage http www sf adfg state ak us FedAidPDFs fds07 44 pdf Burwen D L S J Fleischman and J D Miller
31. with the corresponding fixed values in the DIDSON system Table 5 User configurable parameters in SMC ARIScope data collection software and their corresponding values in DIDSON high frequency identification mode only Parameter ARIS 1200 ARIS 1800 DIDSON LR 1200 DIDSON SV 1800 Transmit es E Length 4us to 100us 4us to 100us 7us 13us 27us 54us 4 5us 9us 18us 36us relative to window length relative to window length Downrange Resolution 3 mm to 10 cm 3 mm to 100 mm 5mm 10mm 20mm 40mm 2 5mm 5mm 10mm 20mm window length samples relative to window length relative to window length Source Level 206 212 dB 200 206 dB re lw Pa at lm re luPa at lm Window Length 2 5m 5m 10m 20m 1 25m 2 5m 5m 10m 7 The downrange resolution 1 e Window Length samples is particularly improved with ARIS Whereas DIDSON was limited to 512 samples to define the downrange resolution ARIS can collect up to 4 000 samples per beam The autofocus feature will be enabled so that the sonar automatically sets the lens focus to the mid range of the selected range window e g for a window length of 10m that started at Sm the focus range would be 15m 5m 2 Table 6 Approximate data storage needs for data from four mainstem ARIS systems at RM 13 75 site Stratum 10 min Hourly Comments file size sample GB size GB 3 ARIS 1800 standard lens Left Bank Offshore nie Collect data for 30 min each hour Stratum 1 5m 25
32. 009 Escapement goal review for Susitna River sockeye salmon 2009 Alaska Department of Fish and Game Fishery Manuscript Series 09 01 Fleischman S J and D L Burwen 2003 Mixture models for species apportionment of hydroacoustic data with echo envelope length as the discriminatory variable ICES Journal of Marine Science Volume 60 Issue 3 Pages 592 598 2 Galbreath P F and P E Barber 2005 Validation of a long range dual frequency identification sonar DIDSON LR for fish passage enumeration in the Methow River Final Report PSC Southern Fund 2004 2005 Project Gaudet D M 1990 Enumeration of migrating salmon populations using fixed location sonar counters Rapports et Proces Verbaux des Reunions Conseil International pour 1 Exploration de la Mer 189 197 209 Gilks W R A Thomas and D J Spiegelhalter 1994 A language and program for complex Bayesian modeling The Statistician 43 169 178 www mrc bsu cam ac uk bugs Accessed 11 2004 Goodman L A 1960 On the exact variance of products Journal ofthe American Statistical Association 55 708 713 Hammarstrom S L and J J Hasbrouck 1999 Estimation of the abundance of late run Chinook salmon in the Kenai River based on exploitation rate and harvest 1997 Alaska Department of Fish and Game Fishery Data Series No 99 8 Anchorage Holmes J A G M Cronkite H J Enzenhofer and T J Mulligan 2006 Accuracy and precision of fish count data f
33. 09 500 cable and SMC wireless bridge that transmits data to right bank for storage and transport to the Soldotna office 92 Right Bank sample scheme O xx 00 xx 10 RB 3 3 8 3m O xx 10 xx 20 RB 8 3 13 3m O xx 20 xx 30 RB 13 3 23 3m O xx 30 xx 40 RB13 3 23 3m O xx 40 xx 50 RB 23 3 33 3m O xx 50 xx 60 RB 23 3 33 3m Left Bank sample scheme Oxx 00 xx 10 LB 3 3 8 3m Oxx 10 xx 20 LB 8 3 13 3m Axx 20 xx 30 LB 13 3 23 3m Oxx 30 xx 40 LB 13 3 23 3m O xx 40 xx 50 LB 23 3 33 3m Oxx 50 xx 60 LB 23 3 33 3m Figure 5 RM 8 6 DIDSON sampling schedules for four range strata on the right top and left bottom banks for 2012 Static IP 128 95 97 227 Static IP 128 95 97 227 DIDSON DIDSON Wireless Bridge L Bank one es One of backup hard backup hard Radio 1 IP 128 95 97 3 drives is drives is Username admin configured to configured to Password smaadmin share data share data i directory directory 4 TB External Hard Drives 4 TB External Hard Drives Wireless Bridge R Bank Radio 1 IP 128 95 97 2 Username admin Password smcadmin On board Ethernet Left Bank Data Collection Computer Right Bank Data Collection Computer Static IP 128 95 97 201 Static IP 128 95 97 200 DFGANCDSFXXXXXX USBjto ethernet USB to ethernet DFGANCDSFXXXXXX IP 192 1 68 1 4 Static IP 192 168 1 3 Username dlburwen Password megaptera USB USB Linksys USB300M Ethe
34. 1 q1 al pa effective 2 a lt pa 2 a for k in 1 5 TL cm 75 k lt TL cm k 75 mu DL1 k lt beta0 beta1 TL cm 75 k DL1 k dnorm mu DL1 k tau DL for i in 1 nfish age i dcat pa effective species i 1 3 mefl mm i dnorm mu species i age i _tau species i age i continued 78 Appendix D2 Page 2 of 2 for j in 1 ntgts species2 j dcat ps age2 j dcat pal species2 j 1 3 mefl mm 2 j dnorm mul species2 j age2 j Ltau species2 j age2 j TL2 cm 75 j lt 1 1 mefl mm 2 j 2 10 75 CONVERSION TO TL mu DL2 j lt beta0 beta1 TL2 cm 75 j DL2 j dnorm mu DL2 j tau DL 79 Appendix D3 WinBUGS code for mixture model under fast track data processing protocol Prior distributions in green font likelihoods in blue model beta0 dnorm 75 0 0025 subjective prior sd 20cm beta1 dnorm 1 25 Fsubjective prior sd 0 2 sigma DL dunif 0 20 tau DL lt 1 sigma DL sigma DL ps 1 2 ddirch D species pa 1 1 dbeta B1 B2 theta1 dbeta B3 B4 pa 1 2 lt theta1 1 pa 1 1 pa 1 3 lt 1 pa 1 1 pa 1 2 pa 2 1 dbeta 0 5 0 5 theta2 dbeta 0 5 0 5 pa 2 2 lt theta2 1 pa 2 1 pa 2 3 lt 1 pa 2 1 pa 2 2 n upstr lt n_meas n_small n chin lt ps 1 n upstr p large lt ps 1 1 pa 1 1 n large lt p large n upstr Lsig 1 1 lt 78 Lsig 1 2 lt 70 Lsig 1 3 lt 74 Lsig 2
35. 10m 20m 40m 80m Range Bin Size relative to window length 20mm 40mm 80mm 160mm Pulse Length relative to window length 23us 46us 92us 184us Identification Mode Operating Frequency 1 2 MHz Beamwidth two way 0 5 H by 14 V Number of Beams 48 Beam Spacing 0 3 nominal Extended Range Settings Extended Window Start 0 42m to 26 1m in 0 42m steps Extended Window Length 2 5m 5m 10m 20m Range Bin Size relative to window length 5mm 10mm 20mm 40mm 58 Pulse Length relative to window length 7us 13us 27us 54us Both Modes Max Frame Rate range dependent 2 21 frames s Field of view 29 Remote Focus 1m to Infinity Control amp Data Interface UDP Ethernet Aux Display NTSC Video Max cable length 100 10BaseT 61m 152m 200ft 500ft Max cable length twisted pair Patton Extender 1220m 4000ft Max cable length fiber optics kilometers Power Consumption 25 Watts typical Weight in Air 7 9 kg 17 4 Ib Weight in Sea Water 1 0 kg 2 2 Ib Dimensions 31 0cm x 20 6cm x 17 1cm Topside PC Requirements Windows XP Vista 7 Ethernet Optional NTSC video monitor 59 APPENDIX B PROCEDURES FOR DAILY DIDSON PROCESSING 60 Appendix B1 Steps for daily DIDSON processing using batch files DAILY DIDSON PROCESSING STEPS Kenai River Chinook Sonar RM 9 Last updated by Debby Burwen 4 20 2012 e Raw Data Transfer and Storage Edit then run didsondownload bat located in O didson12 Batch to move s
36. 14 where 4 1s the AR 1 coefficient the v are model residuals Vai ln ln g 2 15 and the s are independently and normally distributed process errors with white noise variance o Parameters qx and o will be estimated from the data WinBUGS code for the daily abundance model is in Appendix D4 12 II OPERATIONS AT RIVER MILE 13 7 RM 14 RM 14 OBJECTIVES This portion of the study involves deploying imaging sonar and estimating fish passage at a site above tidal influence at RM 13 7 Unlike operations at RM 8 6 where Chinook salmon passage estimates are generated solely for the midsection of river this project seeks to provide estimates for virtually the entire river cross section The primary objective is as follows Estimate weekly and seasonal early and late run upstream passage of Chinook salmon at river mile 13 7 of the Kenai River such that the seasonal estimate 1s within 10 of the true value 95 of the time This estimate will be based on fitting a mixture model to ARIS fish length measurements Because many operational details remain to be worked out at this site any inseason estimates will be preliminary internal quantities only and they may not be available until many days or weeks after data collection RM 14 STUDY DESIGN STUDY SITE This study will be conducted at RM 13 7 on the Kenai River Chinook Sonar Site Figure 1 Figure 14 This location was identified during surveys con
37. 2010 Accuracy and precision of manual fish length measurements from DIDSON sonar images Transactions of the American Fisheries Society 139 1306 1314 Carroll H C and B C McIntosh 2008 Sonar estimation of salmon passage in the Yukon River near Pilot Station 2006 Alaska Department of Fish and Game Fishery Data Series 08 65 Cochran W G 1977 Sampling techniques 3rd edition John Wiley and Sons New York Cronkite G M H J Enzenhofer T Ridley J Holmes J Lilja and K Benner 2006 Use of high frequency imaging sonar to estimate adult sockeye salmon escapement in the Horsefly River Canadian Technical Report of Fisheries and Aquatic Sciences 2647 Daum D W and B M Osborne 1998 Use of fixed location split beam sonar to describe temporal and spatial patterns of adult fall chum salmon migration in the Chandalar River Alaska North American Journal of Fisheries Management 18 477 486 Eskelin A 2010 Chinook salmon creel survey and inriver gillnetting study lower Kenai River Alaska 2007 Alaska Department of Fish and Game Fishery Data Series No 10 63 Anchorage http www sf adfg state ak us FedAidpdfs FDS 10 63 pdf Enzenhoffer H J N Olsen and T J Mulligan 1998 Fixed location riverine hydroacoustics as a method of enumerating migrating adult Pacific salmon comparison of split beam acoustics vs visual counting Aquatic Living Resources 11 2 61 74 Fair L F M T Willette and J Erickson 2
38. 2202 222072 22 DDD II 2522 Ad 1 115 LS mefl mm c 705 820 920 935 430 440 450 460 460 500 500 510 510 510 520 530 530 540 540 550 550 550 550 550 560 560 560 560 560 560 560 560 560 560 570 570 570 570 570 570 570 570 570 570 570 570 570 570 575 580 580 580 580 580 580 580 580 580 580 580 580 580 580 580 580 580 580 585 590 590 590 590 590 590 590 590 590 590 590 590 590 600 600 600 600 600 600 600 600 600 600 600 600 600 600 600 600 605 605 610 610 610 610 610 610 610 610 610 610 610 610 610 620 620 620 620 620 620 620 620 620 620 620 620 620 620 620 630 630 630 630 630 630 640 640 640 640 640 850 970 980 885 930 935 960 430 480 480 5 10 530 540 540 540 550 550 560 560 560 570 570 570 570 570 570 570 570 570 580 580 580 580 580 580 580 585 590 590 590 590 590 590 590 590 595 600 600 600 600 600 600 600 600 600 600 610 610 610 610 610 610 610 610 620 620 620 620 620 620 620 620 620 630 630 640 640 640 640 660 660 735 960 1125 1150 440 450 450 490 500 510 520 520 530 530 540 540 540 550 560 560 560 560 560 560 570 570 570 570 570 570 570 580 580 580 580 580 590 590 590 590 590 590 590 590 590 600 600 600 600 600 6 10 610 610 610 620 620 620 620 620 620 630 630 640 685 765 870 935 1000 1005 1075 610 740 750 885 905 1145 DL2 c 57 60 52 61 54 58 49 67 55 52 102 62 70 65 66 63 98 68 66 95 53 55 48 67 116 60 8 59 57 56 59 71 55 45 85 50 66 69 52 62 49 54 54 61 53 49 59 68 59 62 65 65 66 58 57 112 69 62 58 66 61 65 64 57 59 59 58 58 64 65 55 6
39. 60 MHz DIDSON LR or ARIS 1200 at 0 7 80 28 14 48 0 60 0 60 MHz DIDSON LR or ARIS 1200 at 1 2 30 15 EN 48 0 27 0 30 MHz high resolution lens DIDSON LR or ARIS 1200 at 0 7 100 15 3 48 0 33 0 60 MHz high resolution lens Actual range will vary depending on site and water characteristics Beam width values are for two way transmission at the 3 dB points Values for beam spacing and beam width are approximate Beam widths are slightly wider near the edges of the beam and the beam spacing is slightly narrower Conversely beams are slightly narrower near the center of the beam and the beam spacing is slightly wider e g the center beam spacing is closer to 34 and the beam width is 27 for a DIDSONSS at 1 8 MHz Bill Hanot Sound Metrics Corporation personal communication Nonlinear corrections are applied by the manufacturer in software to correct for these effects in the DIDSON standard but not the high resolution lens Nonlinear corrections are applied in software to correct for these effects in both the ARIS standard and high resolution lens 53 0 2 Figure A1 1 Diagram showing the horizontal plane of a DIDSON LR or ARIS 1200 with a high resolution lens The overall horizontal beam width of 15 is comprised of 48 sub beams with approximately 0 3 beam widths Note that because sub beams grow wider with range fish at close range are better resolved than fish at far range Adapted from Burwe
40. 7 64 59 84 52 63 55 67 60 46 52 59 65 51 58 55 64 63 48 60 50 58 64 62 58 61 61 52 57 59 44 53 48 66 51 45 58 60 50 55 45 64 59 61 61 52 60 57 49 83 51 41 53 61 69 64 73 65 51 49 63 60 64 51 60 55 98 59 57 62 60 62 61 58 63 62 46 65 43 62 43 53 51 65 59 62 63 61 52 57 62 63 69 65 67 61 80 81 55 67 50 56 65 69 46 57 63 52 57 60 59 48 42 65 49 65 50 67 59 59 69 68 55 59 55 65 61 53 59 56 63 57 42 62 50 57 48 53 54 48 65 55 63 63 56 66 51 49 43 52 47 58 65 63 45 80 90 57 47 60 52 40 93 59 110 57 56 63 48 50 85 93 57 102 58 63 62 67 63 69 57 52 87 95 48 63 58 88 56 9 1 49 100 66 60 40 49 56 54 64 55 59 61 90 81 63 75 62 90 51 58 67 104 60 64 47 60 49 60 51 60 40 57 44 58 37 59 51 63 59 57 54 54 51 51 56 51 65 58 45 56 56 68 87 45 63 108 59 63 57 55 58 57 60 59 72 59 53 60 54 57 65 55 54 63 62 46 60 58 54 61 62 59 58 58 67 52 52 52 47 49 59 52 60 56 47 50 50 63 60 58 54 54 43 45 54 65 52 55 56 45 67 50 59 53 51 47 41 49 45 61 58 49 60 52 51 61 50 47 50 65 62 59 61 62 71 61 57 61 64 69 47 53 50 98 59 63 43 50 51 56 52 62 97 62 67 95 59 55 55 60 61 48 57 55 58 50 48 60 60 56 45 57 67 59 59 61 61 72 61 70 69 57 58 52 84 52 46 57 60 57 64 61 58 59 59 48 53 65 68 65 66 65 68 64 64 65 66 74 65 57 62 61 57 53 50 51 50 67 64 55 54 59 60 52 49 61 58 69 61 62 62 108 48 63 64 53 50 56 64 66 62 92 55 56 56 62 57 40 54 61 53 58 59 68 58 57 56 66 48 45 49 46 45 63 63 63 61 60 68 55 101 61 61 60 DL3 c NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA
41. ARIS data for the purpose of generating daily estimates of Chinook salmon passage Primary focus is river mile 13 75 sonar site Work dates April 1 December 31 2013 Michael Friedrich Fishery Biologist I Alaska Department of Fish and Game Assist Project Leader with all aspects of DIDSON ARIS deployment operation and data analysis Take lead role in developing protocol and a comprehensive user manual for processing DIDSON data for the purpose of generating daily estimates of Chinook salmon passage Primary focus is river mile 8 6 sonar site Work dates May 1 August 15 2013 Kara Bethune Fishery Biologist I Alaska Department of Fish and Game Assist Project Leader with all aspects of DIDSON ARIS deployment operation and data analysis Assist in developing a comprehensive user manual for processing DIDSON ARIS data for the purpose of generating daily estimates of Chinook salmon passage Work dates July 15 August 10 to assist with processing data during the peak of the late run and at other times as needed to replace other employees requiring leave Mike Hopp Fish and Wildlife Technician III Alaska Department of Fish and Game On site Crew Leader Supervise provide training and oversee work quality of three technicians based at a semi remote field camp Assist Project Leader with all aspects of DIDSON ARIS deployment operation and data processing Develop batch files and other user interface programs as needed Work da
42. C Control and Display Software Version 5 25 28 aa 68 Appendix D1 Mixture model for estimating species composition of migrating fiSh coocccccncnnnnnnnnnnnnnnnnnnnono 73 Appendix D2 WinBUGS code for mixture model under standard data processing protocol Prior distributions in green font likelihoods in blue ooococcccnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnonos 78 Appendix D3 WinBUGS code for mixture model under fast track data processing protocol Prior distributions in green font likelihoods in blue oooococcnnnnnnnnnnononnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnss 80 Appendix D4 OpenBUGS code for daily abundance model to be used when it 1s not possible to generate Chinook salmon passage estimates with DL mixture model Prior distributions in green font oo AAA o no aujot 82 Appendix D5 WinBUGS code for hierarchical age composition model Posteriors distributions from fitting this model provide prior distributions for DL mixture model Prior distributions in green font TRO OCS bla end 83 Appendix D6 Example WinBUGS data under Fast Track sampling protocol oocccccccnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnos 84 Appendix El Instructions and settings for manual length measurements form ARIS images in 2013 using SING AURIS Bist Sot Wale Version da iia 86 Appendix F 1 Additional details regarding river mile 13 7 sonar site
43. Data Storage and Managemen ss 7 PishrenolinandDir cionolli v lksssbawn ds dd dd 9 RM9D ATRAN LVS Si tic 9 RMeo Midriver salmon Passage Estima rra 9 RM 9 Midriver Chinook Salmon Passage EStIMAteS occccccccccncncncnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnonnnennninannnns 10 RMe9 Midriver Large Fish Passage ES 10 Net Apportioned RM 9 Midriver Chinook Salmon Passage EStIMAtES ooooooononononononoooonnnnnnnnnnnnnnnnnnnnnnononnnnnnnnnnnnnnonoss 10 Modinednons under EE Protocolar didas 11 Moditications under LEC Proton as a o 11 mputaton o Missibe Di 11 Dal Abundance Modes ask odd s s 12 Ii OPERATIONS AT RIVER MILE 137 RM Tota 13 RM 4 OB JHE CPN TES caen ss to adas 13 RMA4ASTUDE DESIGN ao 13 SHUG Rt LE sends ved ERE ST etme da dei da on a dj ERR 13 Sampin Proce dres GRR RR RP ne ira CR a CR RR o a ee 14 RMA DATA COLLECTION ds 16 so ar Equipmentand Contrato da 16 A o AAA A PI USE UEP as s Se CU SURE CPT Ce RC ee 17 Data Sorice and Mana cnn ti aj 19 Pasa ener ang Direction otra dia 20 RNETEDATA ANALY St da 20 RMA Salmon Passade Estates a i mini nesi 20 RM 14 Chinook Salmon Passage EStimate Sei 20 RMA Lares Fish Passage EStdt o i sm oi ds 20 11 Net Apportioned RM 14 Chinook Salmon Passage EStIMAtesS ooooonnnnnoooooonoononnnnnnnonnnnnnnnnnnnnnnnnnnno nono non nono non nro nnnnnnnnoss 20 SCHEDULE AND DELI ERABLE Ss 22 KESPONSIBIEJI IE eeseossenadess odd ds dod 23 B DUFIS i sesosaddieodbeno
44. Model SMC2891 AG IP address 128 95 97 2 Laptop Subnet Mask 255 255 255 0 Media Type 10 Mb sec half Duplex Gateway 128 95 97 227 DIDSON IP IP address 128 95 97 200 Unit MAC Address 00 22 2D 22 42 F9 Subnet Mask 255 255 0 0 Radio A MAC Address 00 22 2D 22 40 AO DIDSON SMC Bridge Right Bank Cat 5 straight through Ethernet cable 500 Cable Newest cable Media converter is not required BUY it can only handle 10 MBit sec data transmission so a 10Mbit hub s inserted between the Topsidg Box and SMC Bridge SMC Wireless Bridge Model SMC2891 AG IP address 128 95 97 4 Subnet Mask 255 255 255 0 SMC Bridge Gateway 128 95 97 227 DIDSON IP Left Bank Unit MAC Address 00 22 2D 22 43 65 Radio A MAC Address 00 22 2D 22 40 30 500 DIDSON cable Ethernet or Coax to Bridge Cat 5 straight J through Ethernet cable Ethernet The weatherproof tote on left bank see photo below holds these components The sonar cable power cord and wireless bridge cable enter the tote through a drilled hole The wireless bridge media power injector module the 10 MBit hub and DIDSON topside box require power via the battery backup The SMC bridge is weather proof D Mbit Hub to convert 10 Mbit to 100 Mbit Power Battery Backup APC 500 or greater To Generator Power ZA lo au IRR U Figure 4 RM 8 6 left bank DIDSON configuration with new since 20
45. NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA N A NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA N A NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA 84 APPENDIX E INSTRUCTIONS AND SETTINGS USED FOR MANUAL FISH LENGTH MEASUREMENTS FROM ARIS IMAGES USING ARISFISH SOFTWARE VERSION 1 5 85 Appendix El Instructions and settings for manual length measurements form ARIS images in 2013 using SMC ARISFish software Version 1 5 a To set Global Settings after a new installation of ARISFish 1 Open ARISFish global settings and ensure you have the following settings Global Application Settings Apply brightness adjustments to the Master image Enable smoothing Y Show range labels Show minor grid lines when the grid ts enabled Show the focus distance indicator Units of measure Meters Celsius Y Clear Settings Files 2 Count Fish Settinas V Display Measured Lengths micusurcu pon vengths will be displayed on the Echogram Export Count File in csv Format Fish count data and column headers only Auto select fish for measurement on mark entry Measure zoomed fish in Rectangular mode Exit measure mode on fish length
46. Puts red marker on fish and automatically activates the movie showing the fish bounded by range arcs see Figure E1 1 below Ze lt Right Click Drag gt on movie image to zoom in for measurement 3 Press lt space bar gt to pause movie 4 Use lt right arrow gt and lt left arrow gt to step through movie one frame at a time to find a frame that displays the entire fish length well see section d below for selecting optimal images 5 lt Left Click Drag gt if necessary to center movie window prior to measuring 6 lt left click gt on the fish snout and continue to lt left click gt along the midline of the fish to create a segmented measurement The segments should follow the midline of the body of the fish ending with the tail J Select lt f gt key to add measurement to the txt file fish it you will see measurement in red lt Left Click gt on echogram inside mark if you want to delete measurement and start over 8 Select lt V gt key to unzoom movie window not necessary if you have another fish nearby you want to measure 9 Next fish repeat steps 1 8 or 10 Left Click gt on Master Echogram to advance to new echogram section or 11 lt Alt gt lt Right Arrow gt to advance to next file Hot keys lt e gt to save all echogram measurements to file lt f gt to fish it to accept the measurement and display it on the echogram lt U gt to undo the last segment lt d gt to dele
47. Regional Operational Plan SF 2A 2013 23 Kenai River Chinook Salmon Sonar Assessment by Debby Burwen Jim Miller and Steve Fleischman February 2014 Alaska Department of Fish and Game Divisions of Sport Fish and Commercial Fisheries Symbols and Abbreviations The following symbols and abbreviations and others approved for the Systeme International d Unit s SI are used without definition in the following reports by the Divisions of Sport Fish and of Commercial Fisheries Fishery Manuscripts Fishery Data Series Reports Fishery Management Reports and Special Publications All others including deviations from definitions listed below are noted in the text at first mention as well as in the titles or footnotes of tables and in figure or figure captions ights and measures metric centimeter deciliter gram hectare kilogram kilometer liter meter milliliter millimeter Weights and measures English cubic feet per second foot gallon inch mile nautical mile ounce pound quart yard Time and temperature day degrees Celsius degrees Fahrenheit degrees kelvin hour minute second Physics and chemistry all atomic symbols alternating current ampere calorie direct current hertz horsepower hydrogen ion activity negative log of parts per million parts per thousand volts watts cm dL ha kg km mL ft s ft gal in nmi OZ lb qt F min ppm
48. SITION The transmit power of the DIDSON sonar is fixed and the maximum receiver gain 40 dB will be used during all data collection The autofocus feature will be enabled so that the sonar automatically sets the lens focus to the mid range of the selected display window DATA STORAGE AND MANAGEMENT Procedures for processing the large volume of data generated daily by the DIDSON systems were developed during the 2010 2012 seasons and will continue to be refined and expanded upon in 2013 Stepwise procedures for daily processing of DIDSON data are given in Table 2 Components of the DIDSON sonar system to be used at river mile 8 6 in 2013 System Component Description Sounder DIDSON LR operating at 1 2 MHz right and left banks Lens Large Lens Assembly with 3 x15 beam pattern right and left banks Data Collection Computer Dell Latitude E6500 laptop computer one for each bank Wireless Bridge For wireless transmission of DIDSON data from left to right bank Model SMC XMC2891 AG Remote Pan and Tilt Aiming Unit Right Bank ROS Model PT25 Pan and Tilt Left Bank SMC Model X2 Rotator Remote Pan and Tilt Aiming Right Bank ROS Model PTC 1 Remote Pan and Tilt Controller Controller Left Bank Controlled through SMC DIDSON software Remote Depth Sensor Right Bank JASCO Model AIM 2000 remote attitude sensor The process of transferring analyzing summarizing and archiving 30 Gigabytes of DIDSON data generated each day is facilitated b
49. a from each sonar for a total of 5 Figure 20 and data will be written directly to one of two external hard drives assigned to that computer sonar Figure 17 The hard drives will be swapped out once per day and transported back to the Soldotna ADF amp G office In the Soldotna office data will be uploaded to a 24 TB Buffalo TeraStation where it can be shared with up to 15 users through a 1 GB Ethernet network 1 e through a 16 port 1 GB Ethernet switch connected to computers with 1 GB Ethernet cards The process of transferring analyzing summarizing and archiving the large volume of ARIS data generated each day will be developed during the 2013 season Batch files similar to those used to handle DIDSON data from the RM 8 6 site will need to be developed These batch files are essential for minimizing errors during data transfer and free personnel to perform other tasks during this time consuming process Remote Access to Sonar Site Because the RM13 7 site will not be manned 24 7 in 2013 a system to remotely access and determine the status of the data collection computers and or sonars will be implemented if possible In the sonar tent an AT amp T MiFi mobile hotspot will be connected to an 8 port wireless router Figure 17 that will network the five data collection computers Each computer will have GoToMyPc installed for remote access Bill Hanot of SMC will be onsite May 13 17 to assist with initial setup and installation of the ARIS
50. agonally dragging a focal window Use forward reverse arrow keys to advance through frames for optimum image measurement Note that direction of travel is logged based on how you measure the target head to tail Press lt F gt to log target Note that echogram will now indicate target length and direction of travel Blue marker downstream Yellow Upstream 69 e Some targets may hold linger in the beam and should not be tracked unless you deem them to have made sufficient progress up or down stream Significant over estimation of fish could occur if everything in the beam were tracked measured without regard for upstream progress e Some targets may glance the corners of the beam if you feel these targets do not provide sufficient direction of travel information or that they may be dipping in an out of the beam throughout the sequence use your discretion at tracking When Chinook fall into this marginal category give them closer attention than smaller targets as missing a small target is less important than missing a KING e Use all the tools at your disposal the echogram provides a roadmap to not only where targets occur in the sequence but often gives strong clues to what the target species is Trace intensity length and tail beat amplitude frequency give you information that comes in handy before you toggle to video mode Video mode and the video loop allow you to better sort out number behavior and location of targets in especially
51. aily abundance model described below may be used postseason to obtain an approximate reconstruction of upstream midriver Chinook salmon passage z as needed IMPUTATION OF MISSING DATA In the event that DIDSON 1s functional for one set of spatial strata but not others 1t may be necessary to estimate the passage on the non functional set s from passage on the functional set s with a ratio estimator Ysi R Ysi gt 1 1 11 where the estimated bank to bank ratio Riss for day i and tide stage t is calculated by pooling counts from all hours during the previous 2 or more days to ensure adequate sample size with tide stage t gt Vsi 2 j T gt Y tio A jeJ jel R EN lt NN O 12 ae mr 2 JEJ jel DAILY ABUNDANCE MODEL Postseason daily upstream midriver Chinook salmon passage z will be reconstructed as needed using the relationship of existing daily Chinook salmon abundance estimates z to daily large fish abundance estimates x and of z to daily catch rates of Chinook salmon in the inriver netting project r Under this model each index x and r generically denoted comprises an independent measure of the relative abundance of Chinook salmon on day i La 92 13 where q 1s the mean ratio of index to true mid river abundance z To allow for a non stationary relationship between each index and true abundance an AR 1 error term will be specified In 7 In q z PV i vi
52. ally Estimate weekly and seasonal early and late run midriver upstream passage of Chinook salmon at river mile 8 6 of the Kenai River such that the seasonal estimate 1s within 10 of the true value 95 of the time This estimate will be based on fitting a mixture model to DIDSON fish length measurements Daily estimates of large fish upstream passage will also be produced based on a DIDSON length threshold In addition a daily net apportioned estimate of Chinook salmon abundance will be produced based on paired netting and DIDSON data Daily estimates of fish passage will be produced for 16 May to 10 August If daily estimates comprise less than 1 of cumulative passage for three consecutive days operations may be suspended before 10 August RM 9 STUDY DESIGN STUDY SITE This portion of the study will be conducted at the existing Kenai River Chinook sonar site an established acoustic monitoring site for Chinook salmon located 14 km 8 6 m1 from the mouth of the Kenai River and operated by Alaska Department of Fish and Game ADF amp G Figure 1 The river is approximately 125 m wide at the site with water depth at mid channel varying from 3 to 8 m due to the strong tidal influence The Kenai River is glacially fed and 1s generally cool and highly turbid Water temperatures during the summer range from 10 C to 15 C and Secchi disk readings from 0 3 to 1 1 m The bottom substrate on the right bank is comprised of fine glacia
53. aragraph of WinBUGS code in Appendix D3 By default data will be stratified by day 1 e daily data from different spatial temporal strata will be pooled and the above model fit to the pooled data Equation 5 Sample size limitations may occasionally necessitate Preliminary work has shown that DIDSON length measurements of tethered fish data are not necessarily representative of measurements from free swimming fish Therefore inclusion of the entire tethered fish data set can bias mixture model results Fortunately only weak prior information about the regression parameters is required in order to estimate species composition with the DIDSON length mixture model and this can be supplied by using only a small subset n 5 of tethered fish data which does not cause substantial bias in the species composition estimates 74 pooling the data across more than one day Changes will not be made to stratification structure during the season Situations that would warrant a post season revision to the estimates include 1 sampling fraction differing greatly among spatial temporal strata or 2 evidence that the slope and intercept parameters differ by stratum In such cases the data will be divided into appropriate groups before analysis and spatial and temporal expansions will be specific to each group 30 50 70 90 110 130 Fish Length cm e e o 0o 0 F e e o e o e o o e o e s j i 30 50 70 90 110
54. ata are used for the length at age means u and standard deviations t Prior information about regression parameters Bo B1 and o is supplied by embedding an analysis of a subset of tethered fish data in the mixture model Figure D1 2 WinBUGS uses Markov chain Monte Carlo methods to sample from the joint posterior distribution of all unknown quantities in the model We will start at least two Markov chains for each run and monitor Gelman Rubin statistics to assess convergence Burn in periods of 10 000 or more samples will be used Samples will be thinned 10 to 1 and at least 10 000 samples per chain will be retained The end product of a Bayesian analysis is the joint posterior probability distribution of all unknowns in the model For point estimates the posterior mean will be used The posterior standard deviation will be presented as an analogue to the standard error of an estimate from a frequentist non Bayesian statistical analysis The mixture model will be fit to daily DIDSON length data but will utilize 7 days of netting data ending on the date of the current DIDSON lengths During times when it is impractical to measure every fish recorded by the DIDSON a Fast Track sampling protocol will be adopted Table 1 and fish measuring less than 75 cm DL will be counted but not measured These fish will be modeled accordingly as having come from a left censored sample The model for censored observations 1s specified in the last p
55. barriers to even wider use of sonar assessment has been the need to estimate the number of spawning salmon separately by species Apportioning sonar counts by species often requires separate intensive sampling programs such as netting programs Bromaghin 2005 Carroll and McIntosh 2008 or fish wheel programs Fair et al 2009 that are costly to implement and subject to biases that can be difficult to resolve Since 1987 ADF amp G has used dual beam 1987 1994 and split beam 1995 2011 side looking sonar technology to estimate Chinook Oncorhynchus tshawytscha passage in the Kenai River Since the project s inception acoustic size average strength of returning echoes and range distance from shore thresholds have been used to differentiate between sockeye O nerka and Chinook salmon These criteria are based on the premise that sockeye salmon are smaller and migrate primarily near shore whereas Chinook salmon are larger and tend to migrate up the middle of the river However studies have shown that these criteria can lead to inaccurate estimates Burwen et al 1998 Hammarstrom and Hasbrouck 1999 Extensive research has been conducted at the Kenai River Chinook sonar site toward improving our ability to identify species from acoustic data Burwen and Fleischman 1998 Burwen et al 2003 Muller et al 2010 Most recently ADF amp G evaluated the potential for dual frequency identification sonar DIDSON to provide improved discrimination of larg
56. ble to generate Chinook salmon passage estimates with DL mixture model Prior distributions in green font likelihoods in blue model q gt80 dnorm 0 1 0E 6 1 0 10 tau log gt80 dgamma 0 001 0 001 phi gt80 dnorm 0 1 0E 4 I 8 8 log resid gt80 0 dnorm 0 4 I 3 3 sigma gt80 lt 1 sart tau log gt80 g ncpu dnorm 0 1 0E 6 I 0 1 tau log ncpu dgamma 0 001 0 001 phi ncpu dnorm 0 1 0E 4 1 8 8 log resid ncpu 0 dnorm 0 4 I 3 3 sigma ncpu lt 1 sqrt tau log ncpu mean log N dnorm 0 1 0E 12 tau log N dgamma 0 01 0 01 N early lt sum N 1 46 N late lt sum N 47 87 for d in 1 D log N d dnorm 0 1 0E 12 I 0 DID d dlnorm log N d tau log DID d gt80 d dinorm log q1Nmean2 d tau log gt80 ncpu d dinorm log q2Nmean2 d tau log ncpu N d lt exp log N d tau log DID d lt 1 log cv DID d cv DID d 1 log q1Nmean1 d lt log q gt80 N d log resid gt80 d lt log gt80 d log q1Nmean1 d log q2Nmean1 d lt log q ncpu N d log resid ncpu d lt log ncpu d log q2Nmean1 d log q1Nmean2 1 lt log q1Nmean1 1 phi gt80 log resid gt80 0 log q2Nmean2 1 lt log q2Nmean1 1 phi ncpu log resid ncpu 0 for d in 2 D log q1Nmean2 d lt log q1Nmean1 d phi gt80 log resid gt80 d 1 log q2Nmean2 d lt log q2Nmean1 d phi ncpu log resid ncpu d 1 82 Appendix D5 WinBUGS code for hierarchical age composition model Posteriors dist
57. complex sequences e Press lt E gt to save your work on each sequence when complete or before you divert to another task Known issues bugs e Use of Fixed Background function Processing gt Background gt Fixed Background sometimes causes the Background Subtraction function to become inoperable when advancing to subsequent files This usually necessitates restarting the DIDSON software Additionally use of this function during tracking 1s necessary if you have reason to scrutinize target s by repeated advancing and retreating frames Under this circumstance background subtraction will make target more and more diffuse unless fixed background function is enabled e During length measurements sometimes the initial mouse click on the target does not work This missed mark just means that you have to first delete your second mark and retry first e Some settings are saved while others may default to undesired values It is wise to periodically check your settings to confirm Oftentimes a change in settings will be apparent during normal tracking and quickly corrected a 99 4 cm b 87 6 cm c 89 8 cm d 97 7 cm e 86 2 cm f 98 6 cm Figure C1 1 Panels a f show the variability in length measurements from DIDSON images of a tethered Chinook salmon during one full tail beat cycle adapted from Burwen et al 2010 70 l l x 2007 07 19_1 M000 IN _ je Elo Edt Yow mge Sonar Grocesing ur Help NSE SK erame Oder
58. ct personnel tor RM 8 6 SE n 26 Table 9 Project personnel Tor MTS SEA A A 26 LIST OF FIGURES Figure Page Figure 1 Map of Kenai River showing location of Chinook salmon sonar sites at river miles 8 6 and 13 7 29 Figure 2 Aerial view and bottom profile at the river mile 8 6 site on the Kenai River with approximate transducer placement and sonar DEAT Coverage used asi ie kesis eng ee cs 30 Figure 3 A DIDSON LR with a high resolution lens photos A and B A silt protection device is used silt sock photo B to protect against silt build up in front of the lens photo C ooooooocoooooooooonononononononononononononononononononoss 31 Figure 4 RM 8 6 left bank DIDSON configuration with new since 2009 500 cable and SMC wireless bridge that transmits data to right bank for storage and transport to the Soldotna OF ICE oooonoonnnonooonooononooonnnnnnnoos 32 Figure 5 RM 8 6 DIDSON sampling schedules for four range strata on the right top and left bottom banks for A P E AE E EEEE EEE E E E ESE E E E E EAE EE E T ETT 33 Figure 6 DIDSON data collection schematic for the RM 8 6 sonar site aa 34 Figure 7 DIDSON data storage configuration and directory structure at RM 8 6 site DIDSON sample files are initially stored on the data collection computer then transferred to the external hard drive using a batch file that also appends an L or R to each filename to indicate which bank the file originated on
59. ctors one of 3 sampling protocols Summarized in Table 1 will guide processing and analysis of rm 8 6 data Standard STD Sampling Protocol Under standard sampling protocol at river mile 8 6 Length will be measured for all salmon shaped fish greater than 40 cm DL Direction of travel will be recorded for each measured fish Fast Track FT Sampling Protocol During the peak of the sockeye salmon run generally mid to late July it can become too time consuming to measure each individual fish given staffing and time constraints associated with generating daily estimates Additionally during periods of peak passage fish often swim in large tightly compacted groups and measurements cannot be uniquely associated with specific fish Figure 10 Under these circumstances at river mile 8 6 the following Fast Track sampling protocol will be practiced Length will be measured for all salmon shaped fish greater than 75 cm DL Length will also be measured for a subset of salmon shaped fish 40 cm lt DL lt 75 cm Either the first F fish or fish passing during the first M minutes of the sampled period will be measured F or M will depend on daily staff time constraints The remaining fish 40 lt DL lt 75 will be marked tallied but not measured To mark fish staff will watch a short section of the DIDSON movie associated with the chart count the number of fish in a group and then make the counted number of marks on the chart
60. d ARIS 1800 Table A1 1 referred to as DIDSON LR HRL ARIS 12000 HRL The high resolution lens has a larger aperture that increases the image resolution over the standard lens by approximately a factor of 2 by reducing the width of the individual beams and spreading them across a narrower field of view Table A1 1 Overall nominal beam dimensions for a DIDSON LR or an ARIS 1200 with a standard lens are approximately 29 in the horizontal axis and 14 in the vertical axis Operating at 1 2 MHz the 29 horizontal axis 1s a radial array of 48 beams that are nominally 0 54 wide and spaced across the array at approximately 0 60 intervals With the addition of the high resolution lens the overall nominal beam dimensions of the DIDSON LR and ARIS 1200 are reduced to approximately 15 in the horizontal axis and 3 in the vertical axis and the 48 individual beams are reduced to approximately 0 3 wide and spaced across the array at approximately 0 3 intervals The combined concentration of horizontal and vertical beam widths also increases the returned signal from a given target by 10dB an effect that increases the maximum range of the sonar over the standard lens A standard lens will be used with the ARIS 1800s deployed near shore at the RM 13 7 site because the wider beam dimensions are preferred for increasing the beam coverage at close range and reducing biases associated with focal resolution at close range see below c Resolution 50
61. day is completed 5 Edit then run package didson bat this program moves the completed target data txt files AND dsamples txt to N didson12 Upload folder to facilitate post processing using database analysis SAS Access Excel 61 Appendix B2 Primer on Batch bat files Batch files are text files containing a series of commands intended to be executed by a command interpreter When a batch file is run the shell program usually COMMAND COM or cmd exe reads the file and executes its commands normally line by line Batch files are useful for running a sequence of executables automatically and are often used to automate tedious processes Before executing a batch file first edit right mouse click lt Edit gt and lt Run gt the file for the correct assignment of Julian ordinal date Date Data Set and drive path Then save the file with the updated settings before running batch Table B2 1 List of Batch files and functions currently used in DIDSON data processing Batch File Name DIDSON bat Transfers DIDSON data from jump drives delivered from camp to the Network Area Storage SAN also referred to as SAN in batch files Creates directory structure for storing each days data by Julian Date bank stratum set Renames each file by inserting an R for right bank files and L for left bank determines bank by sample time in filename Moves data from jump drives to appropriate directory Package didson bat Mov
62. dow Help F pr J Alla la ki koala F r F15l idee har Ma E ip o A E a e Reply with Changes End Review M 111 gt 2 0 EH a h R1 R2 R3 Li L2 L3 137 T 1 1 1 0 0 T 137 1 1 1 1 0 137 2 1 1 1 0 i 137 3 1 1 1 0 0 0 137 4 1 1 1 O D D B 137 5 1 1 1 0 a 137 6 1 1 1 0 137 7 1 1 1 0 0 0 137 8 1 1 1 0 137 g 1 1 1 0 137 10 1 1 1 0 0 0 137 11 1 1 1 O 137 12 1 1 1 0 137 13 1 1 1 0 137 14 1 1 1 0 0 0 137 15 1 1 1 137 16 1 1 1 0 137 17 1 1 1 0 0 0 66 APPENDIX C INSTRUCTIONS FOR MANUAL FISH MEASUREMENTS AT RM 8 6 USING SMC CONTROL AND DISPLAY SOFTWARE VERSION 5 25 67 Appendix C1 Instructions and settings used for manual length measurements from DIDSON images in 2013 at RM 8 6 using SMC Control and Display Software Version 5 25 28 a Parameter setup prior to beginning measurements Step 1 set the number of frames displayed 1 e when right clicking on a fish in echogram mode to display in movie mode from the default of plus minus one second to any number of frames 1 Select lt image gt lt playback gt lt set endpoints gt 2 N Loop on still for N frames 3 Enter the number of frames I suggest 20 30 but you be the judge Step 2 Select lt Processing gt lt Echogram gt lt Use Cluster Data gt if you want to use ALL the beams when creating your Echogram we generally do You can use fewer beams by unch
63. ducted in 2012 and was selected for its favorable physical characteristics for deploying ARIS multibeam technology its accessibility via an adjacent boat launch facility and legal access to property on either bank of the main channel Bathymetric surveys conducted by Aquacoustics Inc on July 9 2012 showed that the section of river at RM 13 7 has a nearly ideal bottom profile for sonar deployment Figure 12 Figure 13 Figure 14 Unlike the RM 8 6 site where only the mid section of the river will be insonified efforts will be made to provide almost complete sonar coverage of the river cross section at the RM 13 7 site Five sonars are required to provide complete coverage a nearhore and offshore sonar on each bank of the mainstem and one sonar on a right bank minor side channel Figure 14 The side channel which has sufficient water for fish passage at higher water levels from approximately mid June through August can be covered by a single sonar combined with a fixed weir on either bank Figured During the early part of the season when the water level 1s low approximately mid May to mid June one sonar on each bank will likely be sufficient to insonify most of the main stem cross section approximately 60 70m But later in the season as water levels rise a second sonar will be deployed on each bank to insonify the nearshore zone and the first 5 10 m in front of the offshore sonars Figure 15 Figure 16 The original now offshore s
64. e Al 2 has twice the down range resolution of the DIDSON image because it was collected at 2 000 samples pixels beam with a 20m Range Window yielding a down range resolution of 1 cm 2 000cm 2 000 pixels compared to a down range resolution of 2 cm for the DIDSON image that was collected at 512 samples with a 10m Range Window 1 000cm 512 The pixels comprising the ARIS image appear less well defined because a smoothing algorithm has been applied Focal resolution in DIDSON and ARIS systems When sizing fish from DIDSON images there can be a bias factor beyond the geometric beam spreading issue depending on the start range and end range of the image window The DIDSON depth of field 1s reduced at closer focusing ranges with the effect that defocused targets will appear smeared in the azimuthal direction The degree of bias is dependent on both the set focus range and the distance of the target from that set focus range In general if the focus 1s set to 5 6m or longer targets will be in good focus from there out to infinity Inside of that range focus will degrade significantly so if the start range 1s lt 2 3m Bill Hanot Sound Metrics Corporation Seattle Washington personal communication For DIDSON focus counts of 0 255 represent the total range of travel of the middle focus lens For the ARIS 1200 1800 which uses the same lens set and has the same focus curves focus counts of 0 1000 represent the total range of travel 0 1
65. e the fish displays its full length e g Panels a d and f in Figure C1 1 In general the best images are obtained when the fish is sinusoidal in shape rather than straight and perfectly perpendicular because the head and tail appear most visible when there 1s curvature to the fish body e g demonstrates the process of measuring a fish using the manual measuring tool The user pauses the DIDSON movie top zooms in on the fish of interest middle and measures the fish length with a segmented line created by mouse clicks along the center axis of the fish bottom The first mouse click is made at the leading edge of the pixel associated with the snout and the final click on the trailing edge of the pixel associated with the tail The software adds the individual segment lengths that are calculated from the pixel coordinates of the DIDSON image Helpful Tips Ensure you can view the entire echogram if you can t see the range hash marks you need to expand the screen further with Toggle Header control and or manually dragging EG window Toggling between movie mode and echogram mode Right mouse click on echogram to see looped movie at same echogram time stamp Spacebar will pause loop Mark targets in EG mode with mouse left click and try to do so somewhere on target track because this also logs target range Right mouse click inside target marker to toggle to movie mode for manual measurement Zoom into target by mouse right click and di
66. e video loop allow you to better sort out number behavior and location of targets in especially complex sequences Press lt E gt to save your work on each sequence when complete or before you divert to another task e Known issues bugs Some settings are saved while others may default to undesired values It is wise to periodically check your settings to confirm 90 f Instructions for selecting optimal images Measurements should be taken from frames where contrast between the fish image and background are high and where the fish displays its full length This can be difficult to determine at times especially when the number of frames from a fish is limited e g close range fish For example panels a b c d g and in Figure E1 3 are images where the fish appears to display its full length and consequently the length estimates are consistently higher than other measurements varying between 95cm 98cm In general the best images are obtained when the fish is sinusoidal in shape rather than straight and or perfectly perpendicular as in panel j because the head and tail appear most visible when there is curvature to the fish body Even when there is curvature to the fish body it is apparent that in some frames the fish body compressed Panel i in Figure E1 3 demonstrates how a fish can measure 99 1 cm dashed line or 88 1 cm solid line depending on whether the user decides to include faint pixels defining the snout
67. ear ProSafe 8 Port Gigabit Cat 6 Switch model GS108 1 GB ethernet Cat 6 gQ J Data2011 2011 05 16 0136 LBank Stratum Seti 6 Search Seti p Organize Burn New folder TEE 0 4 3 didson10 A 192 168 1L101 public N Name Date modified Type Size J 2011 14 5mile prospective site gee a Y 2011 05 16_002000_LHF ddt 2011 05 16_002000_LHf ech J 2011 Project Operational Plan Dell E6500 J 2011 Training Fil es eet faim les pata z 2011 05 16_012000_LHF ech J Batch J W 2011 05 16 022000_LHF ddf Ci Laptops pene 2011 05 16 022000 11462ch opy to op ko iet Y 2011 05 16_032000_LHf ddt El 2011 05 16 032000_LHf ech J 2011 Behind the offshore TX data a p zemus E Y 2011 05 16_042000_LHF ddt ee aaa 2011 05 16 042000_LHF ech mages KES W 2011 05 16_052000_LHF dat p 2011 05 16_052000_LHF ech Y 2011 05 16_062000_LHf ddt 2011 05 16_062000_LHF ech Data Analysis Computer Ill h DAT Static IP for NSD 192 168 1 105 DFGANCDSF 158144 l Didon JJ DATAZE W 2011 05 16 072000 LHF ddt Username FISHSONAR pr eae 2011 05 16 972000 1M ech Password Chinook Y 2011 05 16 082000_LHF daf h ser 2011 05 16 _082000_LHF ech GO i Data2011 2011 05 16D136 FISH Didson_136_DATAZIP Organize Extract all files v le 4 y didson10 192 168 1 101 public N Name 3 R AEAEE JEETS J 2011 14 5mile prospective site J 2011 Establish thresh J 2011 Project
68. easuring consistency 1122 7126 Bill Hanot here to assist with any ARIS or DIDSON issues A report meeting the requirements of ADF amp G s Fishery Data Series will be published This report will provide an overview of the implemented methodology for generating DIDSON based Chinook salmon estimates A draft version will be completed by 4 1 2014 The final report will be completed by 9 1 2014 22 RESPONSIBILITIES List of primary personnel and duties Personnel Duties Personnel Duties Personnel Duties Personnel Duties Personnel Duties Personnel Duties Personnel Jim Miller Fishery Biologist II Alaska Department of Fish and Game General supervision of all aspects of the study Set up and configure DIDSON ARIS sonar system Assist with in season data collection and post season data analysis Primary responsibility for post season report Debby Burwen Fishery Biologist HI Alaska Department of Fish and Game Assist with conducting and supervising all aspects of the study Work with Bull Hanot SMC personnel to develop update needed data processing software Set up and configure DIDSON ARIS sonar system Assist with in season data collection and post season data analysis Assist with testing and evaluating evolving DIDSON ARIS hardware and software Shared responsibility for post season report Mark Jensen Fish and Wildlife Tech IV Alaska Department of Fish and Game Provide expertise
69. ecking this option and selecting the number of beams Step 3 Set up your processing parameters last Icon on right for File Creation as follows v Auto Countfile Name v Binary Count File dat Y New Countfile on Open Y Echogram File ech Step 4 Echogram counts can be reloaded to finish or review at a later time 1f you have checked the Echogram file as follows 1 Select lt File gt lt Open gt then Files of type ech from drop down menu 2 Open desired file 3 The Echogram should reload showing you your previous measurements Or this option will work as long as you saved the dat file as shown above 1 Open the file and bring up your echogram as usual follow instructions below 2 Select lt Processing gt lt Echogram gt lt Import Echogram Counts gt 3 Select the dat file with your saved counts file should reload showing you your previous measurements the filename for the dat file will begin with FC_ Step 5 Make sure lt Image gt lt Configure gt lt Auto Threshold Intensity gt is UNCHECKED This will keep your threshold and intensity settings from changing when you switch between echogram and movie mode Step 6 Uncheck the Display Raw Data toolbar icon first button on left in Combined toolbar If you are in the movie mode and it is displaying the raw image data it is because Display Raw Data is enabled by default b Instructions for manual echogram based length measurements note that these settings ma
70. ect resolution and range capabilities a Frequency DIDSONSs operate at two frequencies a higher frequency that produces higher resolution images and a lower frequency that can detect targets at further ranges but at a reduced image resolution Two DIDSON models are currently available based on different operating frequencies Table A1 1 The short range or standard model DIDSON SV operates at 1 8 MHz to approximately 15 m in range and at 1 1 MHz to approximately 30 m and produces higher resolution images than the long range model The long range model DIDSON LR operates at 1 2 MHz to approximately 30 m in range and at 0 7 MHz to ranges exceeding 100 m but produces images with approximately half the resolution of the DIDSON S see explanation below The two DIDSON LRs used in this study were operated in frequency mode to achieve maximum image resolution Similar to DIDSON ARIS for Adaptive Resolution Imaging Sonar systems operate at two frequencies analogous to the DIDSON frequencies The two ARIS models used on this project ARIS 1800 and ARIS 1200 are essentially updated versions of the DIDSON SV and DIDSON LR models Both ARIS models used in this study were operated in high frequency mode to achieve maximum image resolution b Beam Dimensions and Lens selection The DIDSON LRs and ARIS 1200s used in this study are fitted with high resolution lenses to increase the image resolution to the level achieved by the DIDSON SV an
71. ection EG Width al LA To Right Y Jpstream VW Auto V Clear Counts Frame Region Editor ID Y Show marks 1 5 gt Loop 10 0 gt s Progress Completed 10 Select lt Show EG gt to display the Echogram 87 TZ count Fish TT eee EGG Finish Line CSOT Page P Run Batch 11 ME Finish Line CSOT Z Fish Data e page M Fish 23 Range 9 72 Fr 5162 Bean 1 0 are Retresh EG Run Batch ee anng Length 41 9 Thickness 0 0 More Del Undo Fish 12 Arrange the lt Count Fish gt and lt Fish Data gt windows so your overall display looks similar to z ARIS by Sound Metrics See What Others Can t Le gt Rectangular Palette Deep Blue Flip UR Effects None Grid Measure Geometry T Real time Rate v V 100 amp F un Mevo n Count Fish a Finish Line csoT Page P ERE i ne v x h ia E Fish 0 Range 0 00 Frame 1 Bearing 0 0 Length 0 0 Thickness 0 0 More Del Undo Fish 1 I I 1 1 09 10 10 09 10 20 09 10 30 09 10 40 10 09 11 10 09 11 20 a 09 10 45 8 Up ODown 07 8 Total 6 Measured Frame Tracking OFF 13 Select lt Alt gt lt right arrow gt to advance to the next file when needed all your parameter settings and display configuration should be preserved 14 Now you are ready to start measuring individual fish 88 Instructions for manual fish length measurements using SMC ARISFish software version 1 5 in 2013 1 lt Right Click gt on the fish to be measured
72. ee et ee oe 114 items Offline status Online Offline availability Not available Figure 9 RM 8 6 DIDSON data Storage and file management configuration for Network Storage System in the Soldotna office showing the contents of the fish subdirectory for May 16 2011 The Fish sub directory for each day contains the txt files containing manually measured fish lengths These files are 1 processed locally using SAS to generate daily estimates of fish passage and 2 uploaded to an Anchorage server for further processing by a Biometrician based in Anchorage or Fairbanks 37 E x Y 2011 07 20_074949_HF high density example ddf DIDSON Control and Display V5 25 l File Edit View Image Sonar Processing Aux Help DH SI ere dP rPleae CHAP OOO HCH VFS WORDIE SERENE RS AVY Sonar Controls 243 0 meters Frame Rate o Total Frames o wo pog a E Receiver Gain Window Start Lo w co Window Length a oO co Focus bE EE PERE EE V AutoFreq LF MV AutoRate HF Display Controls MV Reverse Grid MV Smooth 7 Measure Palette LCD2 X Intensity File Position Frame 237 00 00 49 Ped FF n Fir F EF if 4 0 e 07 50 00 07 50 15 07 50 30 07 50 45 07 51 00 07 51 15 2011 07 20 074949 HF high densit 12 71 m 07 52 16 Marked Fish 0 For Help press F1 DEMO MODE MOVIE ONCE Total 9943 First 0 07 51 30 07
73. entry 7 Leave master echogram visible on Hide EG command 2 Enable smoothing is off 3 Display Measured Lengths is on b Set processing parameters for a new set of files for a new day or stratum 1 Select lt Files gt lt Open Recently Viewed gt Open Recently Viewed 2 Navigate to the appropriate directory and open file 3 Set Signal intensity sliders to 0 0 and 40 2 dB or other recommended values for a specific stratum Signal Intensity Histogram 4 Select lt settings cog gt from Filters menu 86 w VY i Count Fish 5 Select lt SMC adaptive background gt SN10 _LB_ PE Platform Motion W Beam Pattern ill Crosstalk Reduction Enable Background Subtraction y Proprietary Sound Metrics Background Subtraction algorithm Updates the background model periodically to respond to background changes an 30 R cm2 6 Select lt Background Subtraction gt icon on Filters Menu Toggle Wa bl Count Fish 7 Select lt Count Fish gt from the Filters Menu to Display the Count Fish Window Y a Von Count Fish 8 Select lt More gt to get expanded options Z Count Fish a MA Finish Line CSOT 4 Pag P Run Batch 9 Increase lt Loop gt Length to at least 8 s set lt upstream gt direction parameter appropriately then select lt Less gt to shrink count Fish window f K i S MA Finish Line CSOT 4 Page D ae PE E 4 Pr upstream Mark Dir
74. epartment s ADA Coordinator can be reached via phone at the following numbers VOICE 907 465 6077 Statewide Telecommunication Device for the Deaf 1 800 478 3648 Juneau TDD 907 465 3646 or FAX 907 465 6078 For information on alternative formats and questions on this publication please contact ADF amp G Division of Sport Fish Research and Technical Seryiges 333 Raspberry Rd Anchorage AK 99518 907 267 2375 SIGNATURE TITLE PAGE Project Title Project leader s Division Region and Area Project Nomenclature Period Covered Kenai River Chinook Salmon Sonar Assessment Debby Burwen James Miller and Steve Fleischman Sport Fish Division Region II Southcentral May 1 2013 April 30 2014 Field Dates 16 May 10 August 2013 Plan Type Category III Approval Title Name Signature Date Project leader Debby Burwen Biometrician Jiaqi Huang Research Coordinator Jack Erickson Regional Supervisor James Hasbrouck TABLE OF CONTENTS Page GSTOR TABLE lo HI LST OF FIGURE S aeo ss SS S ecoRacEoos MI ESTOFATRENDICE uN si ej V PURPOSE izie anai sm j s A d sns l BAG KUR UN Dio il E l OPERATIONS AT RIVER MILE S 6 RM 9 Jrone A A 4 K EJOBJEC FIVE Renar a A A 4 RM9 STUDY DESIGN erence re ren as ee o ee ee ee ps 4 SI DIS Sai Mes RC RO RT RR eR RRC SS ass TE as 4 Sal A Aue ieies vadam E E O EEA 4 RMS DATA COLLEC TION tocata idad 7 Sonar Equipment and CONAM en 7 Di cogu ld mt T nn AA Ss eee eee T
75. er Chinook from smaller species of salmon based on size measurements taken directly from high resolution images of migrating salmon Burwen et al 2007 Based on results of the DIDSON evaluations and due to the inaccuracy of the split beam estimates citations production of split beam estimates was discontinued following the 2011 season and replaced by DIDSON based estimates in 2012 DIDSON is a high definition imaging sonar designed by the University of Washington s Applied Physics Laboratory for military applications such as harbor surveillance and mine identification DIDSON incorporates a lens that provides high resolution images approaching the quality achieved with conventional optics Simmonds and MacLennan 2005 with the added advantage that images can be obtained in dark or turbid water and at farther ranges than is possible with camera technologies Since 2002 ADF amp G has worked closely with Sound Metrics Corporation SMC to adapt this technology to estimating fish passage in rivers Applications of DIDSON systems for monitoring fish populations are becoming more widespread Galbreath and Barber 2005 Holmes et al 2006 Burwen et al 2007 Maxwell and Gove 2007 Mueller et al 2008 and recently DIDSON has been used to obtain information about fish size Cronkite et al 2006 compared the mean length for 2 874 sockeye salmon measured from DIDSON images at an acoustic monitoring site on the Horsefly River with the mean length from
76. es txt files for a given Julian Date AND dsamples txt to a directory N didson12 Upload to facilitate processing and summarizing with SAS Access Excel Resort didson bat Resorts data files into original directories on the SAN if there is an error during uploading Move txt files bat Moves txt files files with measured lengths created by SMC software back into the SAN directories R julian date bank stratum set if necessary e g failure of Package DIDSON batch file below or Echoview related batch files currently not used but available when if Echoview processing is resumed Build directory bat Builds directory structure for Echoview related files Fetch CSOT files bat Copies CSOT files generated by SMC software to the Echoview directory for processing Package csv bat Moves csv files contain auto sizing estimates n created during Echoview processing 62 Appendix B3 DIDSON Control and Display Software Settings and Tools for Tracking using Manual Measurements at river mile 8 6 Open ddf file Check Processing gt Background gt Background Subtraction is ENABLED Check Image gt Configure gt Auto Threshold Intensity is DISABLED Check Image gt Playback gt Set Endpoints to ENABLE Loop on Still for for N frames 30 discretional 20 30 Playback Limits x Total Frames 4164 Start Frame 1 End Frame 4184 Reset Y Loop on Still for N Frames 30 OK Cancel Check Processing gt Echogram gt Use Cluste
77. esterman D L and T M Willette 2006 Upper Cook Inlet Salmon Escapement Studies 2004 Alaska Department of Fish and Game Commercial Fisheries Division Fishery Data Series 06 49 pp 80 82 Westerman D L and T M Willette In prep Upper Cook Inlet Salmon Escapement Studies 2011 Alaska Department of Fish and Game Commercial Fisheries Division Fishery Data Series Wolter K M 1985 Introduction to variance estimation Springer Verlag New York 28 dl February 2014 Divisions of Sh ort Fish and Commer River Sockeye Kill Moose Soldot River ss Skilak Lake z pl nr rr is a j nr ur ur z nr EET EU AEE rr rr M i 3 E z zx Li a z z a a a dl L Figure 1 Map of Kenai River showing location of Chinook salmon sonar sites at river miles 8 6 and 13 7 29 Low Tide Water Level 120 100 80 60 40 20 Right Bank Range m Left Bank Figure 2 Aerial view and bottom profile at the river mile 8 6 site on the Kenai River with approximate transducer placement and sonar beam coverage 30 w yidag ROS PT 25 pan and tilt rotator DIDSON with high resolution lens Figure 3 A DIDSON LR with a high resolution lens photos A and B A silt protection device is used silt sock photo B to protect against silt build up in front of the lens photo C 31 Data Collection Computer on right bank SMC Wireless Bridge
78. etworked with a main computer where data can be reviewed and processed on site see Appendix Table B5 for list of fixed IP addresses Data transfer to the Soldotna office will occur using 32 GB Jump drives Files from individual range strata will vary in size if the ping rate is optimized for each range strata Since the return time for a ping increases with range off shore range strata will require a slower ping rate and files will contain fewer frames In the Soldotna office data will then be transferred to a Network Attached Storage Device Synology DiskStation 1512 with 5 2 TB drives Raid 5 Figure 8 Figure 9 where it can be shared with up to 7 users through a 1 GB Ethernet network 1 e through an 8 port 1 GB Ethernet switch and 1 GB Ethernet cards in each computer FISH LENGTH AND DIRECTION OF TRAVEL Estimates of total length will be made from images using the manual fish measuring feature also included with the SMC DIDSON Control and Display software Collaborative efforts with SMC have resulted in a reasonably efficient method of manually measuring individual fish Detailed instructions for taking manual measurements and the software settings and parameters used are given in APPR E Direction of travel is determined for an individual fish by observation of sufficient frames to classify it as an upstream or downstream migrant RM 9 DATA ANALYSIS Under the standard sampling protocol page 4 also Table 1 abundance estimates wi
79. f the ARIS models used in this study 16 Table 4 Components of the ARIS systems to be used in 2013 System Component Description Sounders 3 ARIS 1200 Left bank mainstem offshore Right bank mainstem offshore Right bank minor channel 2 ARIS 1800 one for each right and left bank nearshore sonars Left bank mainstem nearshore Right bank mainstem nearshore Lens Assembly 2 Standard lens for ARIS 1800 models with 12 x30 beam pattern 3 High resolution lens for ARIS 1200 models with 4 x15 beam pattern Data Collection Computer 5 Dell Latitude E6430 laptop computers one for each sonar Wireless Bridge 3 Wireless Bridge Radio sets either Cisco Aironet 1310 Radiolabs GS2000 or EZ Bridge 5G Lite EZBR 0519 Remote Pan and Tilt 5 Sound Metrics X2 rotator controlled via ARISCOPE software DATA ACQUISITION Unlike DIDSON with ARIS the manufacturers have separated the data collection ARISCope and data processing ARISFish software ARISCope has several data collection parameters that are now user selectable rather than being fixed or limited to a few discrete values as with DIDSON Control and Display Software A consultant from Sound Metrics Corporation will be on site from May 13 17 to assist project personnel with selecting the sampling range intervals and optimizing several parameters for each range interval The parameters that are now selectable and that will be optimized for each range interval are given in Table 5 along
80. f the range Table 3 lists approximate range strata based on river bottom profiles from bathymetric data collected in July 2012 Figure 13 Figure 15 Like DIDSON the ARIS can be programmed to automatically sample each range stratum 15 Table 3 Approximate range strata for sampling at RM 13 7 for 2013 ARIS 1200 LL 10 20m ARIS 1200 LL 20 35m ARIS 1200 LL 10 20m ARIS 1200 LL 20 30m All fish greater than 75 cm will be measured as well as a subset of fish less than 75 cm Because we are likely to see far more sockeye salmon at the RM 13 7 site with near bank to bank coverage we will likely have to make greater use of Fast Track or Large Fish Only sampling protocols described on page 4 RM 14 DATA COLLECTION SONAR EQUIPMENT AND CONFIGURATION At the rm 13 7 site Adaptive Resolution Imaging Sonar ARIS developed by the manufacturers of DIDSON will be used ARIS has several advantages over DIDSON technology including user configurable window lengths variable transmit pulse lengths and increased downrange resolution Additionally ARIS is a sealed system which should negate the need for using a silt excluding enclosure to protect the system from silt buildup inside the lens cavity Figure 3 Components of the ARIS system are listed in Table 4 The theory behind ARIS multibeam technology is similar to that of the DIDSON and is summarized in Appendix A along with a discussion of the features o
81. gmented measurement The segments should follow the midline of the body of the fish ending with the tail Try not to use more than three or four segments to define the fish see section C below 12 lt double left click gt or select lt f gt key to add measurement to file fish it 13 lt right click gt to unzoom 14 lt right click gt to return to Echogram Hot keys l lt e gt to save all echogram measurements to file 2 lt f gt to fish it to accept the measurement and display it on the echogram 3 lt u gt to undo the last segment 4 lt d gt to delete the all segments 5 lt space bar gt to pause in movie mode if this doesn t work click in the black area of the display 6 lt right arrow gt forward direction when you select play or advances frame one at a time if the pause button is on pause button blue square on the toolbar 7 lt left arrow gt opposite of above 8 Left Click Drag to show movie over the selected time Right Click Drag zooms the selected area Table C1 1 Threshold and intensity settings for range strata To adjust these settings use slider bars under the Display Controls to the left side of the Echogram or Movie window P 3 3 8 3m 8 3 13 3m 13 3 23 3m 23 3 33 3m Threshold Intensity c Selecting optimal images to measure Measurements should be taken from frames where contrast between the fish image and background are high and wher
82. he tent will be located in a clearing to alleviate the need to remove trees and other vegetation A small plywood box 3ft x 3ft lined with plastic for spill containment located near the tent will house a small generator A power cable will lead from the generator to the tent Electronic control cables will lead from the tent into the river and attach to each of two ARIS units nearshore and farshore sonars mounted to removable steel o The sonars will be positioned on the river bottom approximately 6 ft and 24 ft from OWH On the right bank main channel site two waterproof totes 3 x 4 will house a battery bank Figure F1 6 and the topside sonar electronics Figures F1 6 to F1 8 A plywood box 3 x 3 lined with plastic for spill containment located near the totes will house a small generator 8 A power cable will lead from the batteries to the tote housing the ARIS electronics Electronic control cables will lead from the topside electronics tote into the river and attach to each of two ARIS units nearshore and farshore sonars mounted to removable steel tripods The sonars will be positioned on the river bottom approximately 6 ft and 45 ft from OWH On the right bank minor channel site a single waterproof tote 3 x4 will house the battery bank Figure F1 10 and topside sonar electronics Electronic control cables will lead from the topside electronics tote into the river and attach to a single ARIS unit mounted o
83. iew Main Channel Left Bank looking downstream Tripod Deployment Submerged steel tripods 6 ft and 24 ft from OHW with ARIS units attached location will be identified Electroniccontrol cables by a marker float on the leading from tent to sonar water s surface unitin the river Figure F1 5 Side view of main channel left bank sonar tripod deployment at the RM 13 7 sonar site 101 K 7 Bridge Off Shore Battery Bank Up to 4 7 AGM L16 6 Volt Batteries 6 W idle 30 W moving max 36 5 W hr 876 W day ARIS 30W 6 Widle 30 W moving max ARIS 30W 36 5 W hr 876 W day Figure F1 6 Proposed schematic for the supplying DC power to the two right bank main channel ARIS systems at RM 13 7 via a battery bank charged by a 2000W generator 102 Top View Main Channel Right Bank looking downstream Walkway Platform Tripod Deployment ADF amp G Sport Fish D Sonar Site Investigation Study River Mile 13 7 KPB Parcel No 055 250 30 Scale Sheet 6 of 11 Date 3 Apr 13 _1in 12ft Submerged steel tripods 76 ft and 45 ft from OHW with ARIS units attached location will be identified by a marker float on the water s surface Ordinary High Water River Flow Figure F1 7 Aerial view of main channel right bank battery bank and sonar deployment at the RM 13 7 sonar site 103 Side View Main Channel Right Bank
84. ite Length data are paired with hydroacoustic data from the same time periods In this version of the analysis we assume no gillnet size selectivity Sockeye and Chinook salmon return from the sea to spawn at several discrete ages We modeled sockeye and Chinook length distributions as three component normal age mixtures fs x Os1 fs1 x Os2 fs2 x Os3 fs3 x D1 3 fc x Oc1 fci Oc2 fc2 x Oc fcs D1 4 where Oca and Og are the proportions of Chinook and sockeye salmon belonging to age component a fsa x N Lsa T sa and D1 5 fca x N HcasT ca D1 6 The overall design 1s therefore a mixture of transformed mixtures That 1s the observed hydroacoustic data are modeled as a two component mixture of y each component of which is transformed from a three component normal mixture of x Bayesian statistical methods will be employed because they provide realistic estimates of uncertainty and the ability to incorporate auxiliary information We will implement the Bayesian mixture model in WinBUGS Bayes Using Gibbs Sampler Gilks et al 1994 Bayesian methods require that prior probability distributions be formulated for all unknowns in the model Species proportions mg and Tc are assigned an uninformative Dirichlet 1 1 prior Age proportions Osa and Oca are assigned informative Dirichlet priors based on a hierarchical analysis of historical data Appendix D5 Likewise informative normal priors based on historical d
85. l mud and on the left bank the substrate is fine gravel with larger cobble The slope from either bank to the thalweg approximately 3 on right bank and 5 5 on left bank is gradual and uniform SAMPLING PROCEDURES At the rm 8 6 site a single DIDSON transducer will be deployed on each bank of the river such that 1t remains submerged at low tide As discharge and water levels rise during the summer tripods may periodically be moved up the river bank Transducers deployed on each bank will operate simultaneously switching between spatial range 1 e distance from transducer strata in 10 minute increments using the schedule in Figure 5 The 4 range strata are 3 3 8 3 8 3 13 3 13 3 23 3 and 23 3 33 3 meters from the transducer The 13 23m and 23 33m strata will be sampled twice hourly however in general only the data Between and greater than 3 m in range from both transducers 4 from the first of the 2 periods will be processed Data from the redundant samples will be archived and processed only in the event that the first sample is not usable Range direction of travel and length cm as measured with a graphical software utility see below will be recorded by fish Details regarding which fish to measure and whether or not to record direction of travel differ depending upon rate of fish passage level of staffing time constraints related to fishery management and fish behavior On a given day depending on the above fa
86. ll be produced as follows RM 9 MIDRIVER SALMON PASSAGE ESTIMATES The number of salmon y of all species exceeding DL 40 cm during day 7 that migrate upstream at RM 8 6 in mid river at least 3 m from the face of each transducer will be estimated as I DD 1 J where 8 dy Didar 2 and y 1s the estimate of passage in stratum s during hour j of day i as follows 60 Ys P Cw 3 sij where Csi number of upstream bound fish greater than 40 cm in stratum s for hour j of day i ts number of minutes usually 10 sampled in stratum s during hour j of day i The sampling variance of the fish passage estimates for stratum s on day i due to systematic sampling in time will be approximated successive difference model Wolter 1985 with adjustments for missing data as 24 M P T PU Y ul 4 MAE sad gt 3 2 20 2 PiP where f is the sampling fraction proportion of time sampled daily often 0 17 and gj is 1 if y exists for hour j of day i or 0 if not RM 9 MIDRIVER CHINOOK SALMON PASSAGE ESTIMATES The estimate of Chinook salmon abundance on day 7 will be calculated by multiplying the fish passage estimate by the estimated proportion of Chinook salmon 7 derived by fitting a DIDSON length mixture model to upstream DIDSON and netting data as described in Appendix D Z y Ti Ci 5 The variance estimate follows Goodman 1960 v r 2 9 var 2 v r v r var 6 The cum
87. looking downstream Walkway and Platform ADF amp G Sport Fish Sonar Site Investigation Study River Mile 13 7 Scal KPB Parcel No 055 250 30 lin Bf Sheet 7 of 11 Date 3 Apr 13 Two wireless radio antennas strapped to tree to transmit data to tent located across the river Small 2000 watt generator housed in a 3 ft x 2 ft Removable elevated wooden box lined with light penetrating plastic for spill containment walkway 3 ft wide and steps 3 ft x 11 ft Battery bank and sonar topside equipment housed in two 4 ft x 3 ft plastic container totes Ordinary High A ad Water A U VA Figure F1 8 Side view of main channel right bank battery bank and walkway deployment at the RM 13 7 sonar site 104 Side View Main Channel Right Bank looking downstream Tripod Deployment Submerged steel tripods 76 ft and 45 ft from OHW with ARIS units attached location will be identified by a marker float on the water s surface Electroniccontrol cables leading from topside totes to sonar units in the river Figure F1 9 Side view of main channel right bank sonar tripod deployment at the RM 13 7 sonar site 105 I 7 1000 W Generator Occasional Use Only Wireless Bridge Back Channel Battery Bank Two AGM L16 6 Volt Batteries 6 W idle 30W i 36 5 W hr 876 W day moving max ARIS X2 Rotator Topside Back Figure F1 10 Proposed system
88. m Left Bank Offshore 1 5 Collect data for 30 min each hour ARIS 1200 Large Lens Stratum 2 25m 35m Left Bank Nearshore 0 Collect data for 60 min out of each hour See comment 1 below 1 ARIS 1200 Large Lens Right Bank Nearshore 0 5 3 0 Collect data for 60 min out of each hour See ARIS 1800 standard lens comment 1 below Right Bank Offshore 1 5 Collect data for 30 min each hour ARIS 1200 Large Lens Stratum 1 Sm 25m Right Bank Offshore 1 5 Collect data for 30 min each hour ARIS 1200 Large Lens Stratum 2 25m 35m Total Hourly AA AA SS Total Daily hourly 24 TO AAA Season Total Daily 87days 6 264 5 056 Or 25 TB for season ARIS 1800 files could be twice as large if collected at same resolution as ARIS 1200 because it has twice as many sub beams 96 Left and right bank near shore sonars will not be deployed until water levels rise mid June so the 25 TB estimate for the season is generous 0 5 0 5 0 5 0 5 0 5 3 0 72 0 3 Assumes we do not need more than two strata for the offshore sonar Does not include the minor channel ARIS these files can be left on the external drives Although ARIS file sizes are bigger than DIDSON files ARIS strata can be longer in range and will not necessarily generate larger files 18 DATA STORAGE AND MANAGEMENT Individual files for each 10 minute sample will vary in size due to different ping rates for each
89. metimes causes the Background Subtraction function to become inoperable when advancing to subsequent files This usually necessitates restarting the DIDSON software Additionally use of this function during tracking is necessary 1f you have reason to scrutinize target s by repeated advancing and retreating frames Under this circumstance background subtraction will make target more and more diffuse unless fixed background function is enabled During length measurements sometimes the initial mouse click on the target does not work This missed mark just means that you have to first delete your second mark and retry first Some settings are saved while others may default to undesired values It is wise to periodically check your settings to confirm Oftentimes a change in settings will be apparent during normal tracking and quickly corrected 65 Appendix B4 Creating and Editing the dsamples txt file for SAS processing Edit the dsamples txt file R upload This file is used by SAS to identify any missing samples and is uploaded each day for SAS database analysis This file is best modified in the file 2012 DIDSON daily summary of data quality xls R spreadsheets then copied into the dsamples txt file in notepad using a copy paste command needs to be a tab delimited file _ are Fa CF LU E Microsoft Excel 2010 DIDSON daily summary of data quality thru 7 21 xls ie la B gt j Hle Edit Format Tools Data Win
90. n a steel approximately 40 from OHW A picket weir will extend approximately 45 from OHW to force fish in front of the sonar Submerged steel tripods will be marked with a marker float on the water s surface Figures F1 5 and F1 9 Additional large red warning buoys will be placed near the offshore sonar tripods on each bank Signage warning boaters to stay offshore of the warning buoys will be posted at all boat launches Figure F1 14 and signage identifying the location of the new sonar project will be will be placed approximately 300 upriver and downriver of the site Looking downstream 96 i I Ps 055 250 23 R Parcel No 055 250 23 055 250 30 i IAEA Sheet 1 of 11 Date 3 Apr 13 ti M V S S l i 1 ie i KPB 055 250 30 vi i LM i i l Kenai Riverbend l Resort Mi 9292 Google S RR 60231106 004 N S917 UN Gay GI Figure F1 1 Aerial map and parcel numbers for RM 13 7 sonar site 97 KPB 055 250 23 i A a vi Y T i IX y ni p E P Cabl 4 poru able gt gt e pe MM Data Cables ELP walkway w stairs Kenai Riverbend Resort Se ga 62912 Google Ethernet Cable Leadingto Wireless Radio Mounted on Tree md Near the West Side of Island ep i 999191 N 1909339039 VS D Figure F1 2 Aerial view of sonar deployment at the RM 13 7 sonar site RS A Tot E Tooke ae River Mile 13 7 KPB Parcel No 055 250
91. n db os ni nO 26 RPEFEREN CESCIEDtessessedd sdd sos ads e e nd do 27 APPENDIX A DIDSON AND ARIS CONFIGURATIONS USED ON KENAI RIVER CHINOOK SONAR PROJTECTS AT RM S 6 AND RWMAS 20 o e S S 49 APPENDIX B PROCEDURES FOR DAILY DIDSON PROCESSING issii a 6 APPENDIX C INSTRUCTIONS FOR MANUAL FISH MEASUREMENTS AT RM 8 6 USING SMC CONTROLAND DISPLAY SORTWARE VERSION S 2 daa das 67 APPENDIX D DIDSON LENGTH MIXTURE MODEL AND ASSOCIATED BUGS PROGRAM CODE ip APPENDIX E INSTRUCTIONS AND SETTINGS USED FOR MANUAL FISH LENGTH MEASUREMENTS FROM ARIS IMAGES USING ARISFISH SOFTWARE VERSION 1 oooccccccnnnoconncnnncccnannos 85 APPENDIX F RIVER MILE 13 7 SITE INVESTIGATION DOCUMENTION coooooooooccnncnonononnnnnnnnncnonnnnnnnnnnnccnnnnnns 95 APPENDIX Gs IP DDRESSES saks coo eee ee 111 LIST OF TABLES Table Page Table 1 Sampling protocols to be applied in 2013 6 Table 2 Components of the DIDSON sonar system to be used at river mile 8 6 in 2013 aa 8 Table 3 Approximate range strata for sampling at RM 13 7 for 2013 16 Table 4 Components of the ARIS systems to be used in 2013 17 Table 5 User configurable parameters in SMC ARIScope data collection software and their corresponding values in DIDSON high frequency identification mode only 17 Table 6 Approximate data storage needs for data from four mainstem ARIS systems at RM 13 75 site 18 Lable 7 2013 timeline and inlestOn issen e eo ausmas 22 Table S Proje
92. n et al 2007 54 Pixel Height y Pixel width Figure A1 2 An enlargement of a tethered Chinook salmon showing the individual pixels that comprise a DIDSON image top contrasted with an ARIS image of a free swimming Chinook salmon bottom 55 thoes Y gt gt au Vi Y l LL Vi gt Q Vi U O LL Vi C u onal U la 0 U C U rr U Y Cc XL a S 5s 8 8 83 8 88 BSS g Oo Me OW NA F M N H un Jad gT JO 000T 01 Q WO g BAG paj9Ae a3ue e10 sdo s sjunes sn30 UOIJISO y sua Range m ARIS Big Lens Focus Curve O s 8 8889855 g 0 FP O A M N H 9AEJ 2101 Jo Wo 0 8 1340 UOHHISOd SU vt CEC CCE CIC coc col CST CLI cor cSt ttl cel ATA TT COL 6 8 EL 9 cs tt ce EG Range m Relationships between focal length and lens position for ARIS 56 Figure A1 3 Appendix A 2 Manufacturer specifications for sonar models ARIS 1200 ARIS 1800 DIDSON SV and DIDSON LR 1 ARIS 1800 Specifications Detection Mode Operating Frequency 1 1 MHz Beamwidth two way 0 5 H by 14 V Source Level average 200 206 dB re 1 uPa at 1 m TBD Nominal Effective Range 35m Identification Mode Operating Frequency 1 8 MHz Beamwidth two way 0 3 H by 14 V Source Level average 200 206 dB re 1 uPa at 1 m TBD Nominal Effective Range 15m Both Modes Number of beams 96 or 48 Beam Spacing 0 3 nominal H
93. not clear whether an ARIS 1800 will have sufficient range capabilities to cover the near shore area on the right bank where the offshore sonar may be positioned 15m from the OHW Appendix Figure F19 Consequently an ARIS 1200 with a high resolution lens will be used in 2013 to insonify the right bank nearshore area Once the project is operational and if 1t seems feasible to use an ARIS 1800 one will be borrowed from another project for a short trial period Sampling for both banks will be controlled by electronics housed in a tent located on the left west bank of the river Figure 14 The ARIS units will be mounted on SMC X2 pan and tilt units for remote aiming in the horizontal and vertical axes Similar to DIDSON deployment at RM 8 6 the sonar and rotator units will be deployed in the river using a tripod style mount Figure 3 In the horizontal plane the sonar will be aimed perpendicular to the flow of the river current to maximize the probability of insonifying migrating salmon from a lateral aspect Internal sensors in the ARIS will provide measurements of compass heading pitch and roll as well as water temperature Communication cables from the left bank ARIS units will feed directly into the left bank Top Side Box and data collection computers Figure 17 On the right bank data from the three ARIS systems will be transmitted via three wireless bridges to three data collection computers on the right bank Figure 17 Figure 18
94. ns for taking manual measurements and the software settings and parameters are given in RM 14 DATA ANALYSIS Abundance estimates at RM 13 7 will follow their counterparts at RM 8 6 and will apply to the entire river cross section RM 14 SALMON PASSAGE ESTIMATES The number of salmon y of all species exceeding DL 40 cm during day 7 that migrate upstream at RM 8 6 in mid river at least 3 m from the face of each transducer will be estimated as RM 14 CHINOOK SALMON PASSAGE ESTIMATES The estimate of Chinook salmon abundance will be calculated by multiplying the RM 13 7 salmon passage estimate by the estimated proportion of Chinook salmon 7 derived by fitting a DIDSON length mixture model to upstream DIDSON and netting data as described in equations 5 8 RM 14 LARGE FISH PASSAGE ESTIMATES The daily estimate of large fish passing RM 13 7 will be obtained with equations 1 4 after substituting C sii for Csi where C s number of upstream bound fish in the ensonified zone exceeding 75 cm in length as measured by the ARIS during tsij NET APPORTIONED RM 14 CHINOOK SALMON PASSAGE ESTIMATES The net apportioned daily estimate of Chinook salmon abundance at RM 13 7 will be obtained following equations 9 and 10 20 Modifications required under FT and LFO sampling protocols described on page 11 also apply to RM 13 7 estimates Missing data from RM 13 7 may require imputation following equations 11 and 12 conducted pos
95. nusable for measuring fish size During the remaining 45 days 15 days early run 30 days late image resolution was sufficiently high to generate daily estimates of Chinook salmon passage Although image resolution was frequently reduced in the far range 23 33 m strata there was little evidence that this seriously impacted the ability to distinguish large from small fish Comparison of paired DIDSON and split beam data revealed that DIDSON provided improved estimates of Chinook salmon not only because it provides more accurate estimates of fish length but because it could also interpret complex fish behavior more accurately than the split beam distinguish seals and other targets that were incorrectly classified as large Chinook salmon by the split beam sonar and detect salmon potentially masked by eulachon schools Miller et al 2012 In 2011 DIDSON was again deployed on both banks and sampled the same area covered by the split beam sonar from May 17 through August 10 2011 Focus issues were resolved and few hardware or software related problems were encountered Quality control procedures for on site data collection and office based data processing were developed and refined Because data from 2010 DIDSON operations indicated that sizeable numbers of Chinook salmon were traveling in the nearshore strata 1 e 3 13m from each transducer a third DIDSON was deployed to insonify the region 10m immediately behind the existing left bank tripod
96. of ARIS data back to the main camp on left bank A wireless bridge transmits data to a data collection computer on left bank for storage and subsequent transport to the Soldotna office cc cccccccccceeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeees 46 Figure 19 Schematic of power requirements for RM 13 7 left bank main camp electronics ooooooooccccnnnnnnnnnnns 47 Figure 20 Schematic of RM 13 7 left bank main camp equipment configuration occccccccnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnninininos 48 1V LIST OF APPENDICES Appendix A 1 DIDSON and ARIS configurations used on Kenai River Chinook sonar projects at RM 8 6 and RM 13 7 2013 including an overview of features that affect resolution and range capabilities 50 Appendix A 2 Manufacturer specifications for sonar models ARIS 1200 ARIS 1800 DIDSON SV and DID ON TR a da a deserted 57 Appendix B1 Steps for daily DIDSON processing using batch files ccoonnnnooococonocnncncnonononananonanonnnnnnnnnnnnnnnnnns 61 Appendix Primer on Batch Dates aia 62 Appendix B3 DIDSON Control and Display Software Settings and Tools for Tracking using Manual Measutements al fiver mile 0 sad 63 Appendix B4 Creating and Editing the dsamples txt file for SAS processing oocoooooooooconcnnnnncccnnnnnonanonononnnnnnnnnnnnnnnnnnns 66 Appendix C1 Instructions and settings used for manual length measurements from DIDSON images in 2013 at RM 8 6 using SM
97. on 20 26 July Substantial numbers of large fish were detected during this limited deployment In 2012 the split beam sonar was discontinued and DIDSON was used to generate inseason Chinook passage estimates Escapement goals based on DIDSON generated estimates of passage were developed Fleischman and McKinley 2013 McKinley and Fleischman 2013 Additionally in 2012 crew conducted a two week evaluation of a new sonar site located above tidal influence and tested the performance of the newest generation of DIDSON technology referred to as Adaptive Resolution Imaging Sonar or ARIS The current site at RM 8 6 was selected primarily because of its suitability for operating a dual beam and subsequently a split beam sonar system which requires a near perfect linear bottom profile over the entire insonified zone or in this case from the near shore region to the thalweg However this site has many disadvantages primarily related to its location within tidal influence such as 1 incomplete coverage of the river due to tidal activity flooding the region behind the transducers 2 milling fish behavior related to tidal flux 3 physical risk to gear by large debris carried by extreme tidal fluxes and lack of legal access to the property on one bank Relocating the site farther upriver could improve ADF amp G s ability to more accurately estimate king salmon passage by minimizing or eliminating these negative factors In 1999 ADF amp G searched for and
98. onars cannot be moved closer to shore as water levels rise because they already insonify the maximum range recommended for operation in high frequency mode approximately 30 35m Append Two different ARIS models will be used to provide optimal coverage of the mainstem cross section Table 4 Fable 3 Figure 16 ARIS 1200 models with high resolution lenses will be used as the offshore sonars because they have the higher range capabilities up to 35m from tests conducted in 2012 needed to insonify the majority of the mainstem river at lower water 13 levels and the offshore region of the mainstem during higher water levels The ARIS 1800 with a standard lens is more advantageous for insonifying close range targets and nearshore areas because it operates at a higher frequency yielding higher resolution and the standard lens also has better focusing capabilities at closer ranges Appendix Figure A1 3 Additionally the wider beam dimensions of the ARIS 1800 without the HRL 14 x28 versus 4 x15 provide better coverage in both vertical and horizontal dimensions at short ranges Finally using sonars with different operating frequencies will allow nearshore and offshore strata to be sampled simultaneously without crosstalk interference if desired In 2013 an ARIS 1800 system will be deployed as the nearshore sonar on the left bank because the offshore sonar will be positioned only 8m from the bank at ordinary high water OHW However it is
99. orizontal Field of View 28 Max frame rate 96 beams 3 15 frames s 6 15 frames sec w 48 beams Minimum Range Start 0 7m Downrange Resolution 3mm to 10cm Transmit Pulse Length 4us to 100us Remote Focus 0 m to max range Power Consumption 15 Watts typical Weight in Air 5 5 kg 12 1 Ib Weight in Water TBD 1 4kg 3 Ib Dimensions 31cm x 17cm x 14cm Depth rating 300m Data Comm Link 100BaseT Ethernet Maximum cable length Ethernet 90m 300 feet 2 ARIS 1200 Specifications Detection Mode Operating Frequency 0 7 MHz Beamwidth two way 0 8 H by 14 V Source Level average 206 212 dB re 1 uPa at 1 m TBD Nominal Effective Range 80m Identification Mode Operating Frequency 1 2 MHz Beamwidth two way 0 5 H by 14 V Source Level average 206 212 dB re 1 uPa at 1 m TBD Nominal Effective Range 25m Both Modes Number of beams 48 Beam Spacing 0 3 nominal Horizontal Field of View 28 Max frame rate range dependent 2 5 15 frames s Minimum Range Start 0 7m Downrange Resolution 3mm to 10cm Transmit Pulse Length 4us to 100us Remote Focus 0 m to max range Power Consumption 15 Watts typical Weight in Air 5 5 kg 12 1 Ib Weight in Water TBD 1 4kg 3 Ib Dimensions 31cm x 17cm x 14cm Depth rating 300m 57 Data Comm Link 100BaseT Ethernet Maximum cable length Ethernet 90m 300 feet DIDSON SV Specifications Detection Mode Operating Frequency 1 1 MHz Beamwidth two way 0 4 H by 14
100. ort data from thumb drives to processing directory on the Network Storage Device l 2 3 Change Julian date Change the Calendar date Change the drive label s for the jump drives if needed you will see the drive label when you plug the drive in e Manual Target Measurement l Using DIDSON Application remember it s free software process ddf files for manual measurement of targets Basic sequence of events Open file create echogram measure targets save file lt e gt deselect echogram advance to next file black right arrow See appendices B3 and D1 for more tracking details and appropriate software settings parameters 3 At the end of a strata s worth of data verify that you have not skipped any files and log targets of interest if you haven t already done so Verify that file sizes are the same a check on sample length Edit the dsamples txt file located in R upload This file is used by SAS to identify any missing samples and is uploaded each day for SAS database analysis This file is best modified in the file 2012 DIDSON daily summary of data quality xls R spreadsheets then copied into the dsamples txt file in notepad using a copy paste command needs to be a tab delimited file Copy dsamples txt to the data directory for that day e g to R Data2012 2012 08 02 JD214 See appendix B4 for more detail on editing the dsamples file and uploading data Once all data for the
101. per unit Figure Al 3 shows the ARIS lens position indicated by the numbers in the range 0 1000 versus focus range for the ARIS High Resolution Large lens There is a non linear relationship of lens position to focus range with short range focus requiring large movements for small increments in focus range and long range focus having small changes in lens position for several meters of change in focus range Also beyond a certain range images are generally in focus Based on the focus curves in Figure A1 3 images are at least 75 in focus starting at 4 m for the standard lens and starting at 7 m for the large lens 32 Table A1 1 Summary of manufacturer specifications for maximum range individual beam dimensions and spacing for DIDSON SV DIDSON LR ARIS 1800 and ARIS 1200 systems at two frequencies with and without the addition of a high resolution lens specifications from Sound Metrics Corporation A more complete summary 1s given in Appendix A2 Maximum Horizontal Vertical Number Individual Individual range m beam beam ofbeams beam beam width width width spacing System DIDSON SV or ARIS 1800 at 1 8 15 28 14 96 0 30 0 30 MHz DIDSON SV or ARIS 1800 at 1 1 30 28 14 48 0 40 0 60 MHz DIDSON SV or ARIS 1800 at 1 8 20 15 3 96 0 17 0 15 MHz high resolution lens DIDSON SV or ARIS 1800 at 1 1 40 15 3 48 0 22 0 30 MHz high resolution lens DIDSON LR or ARIS 1200 at 1 2 23 28 14 48 0 40 0
102. ply Defaults Ensure display controls are properly adjusted unique to each strata Strata 1 Intensity 50 Threshold 11 Strata 2 Intensity 50 Threshold 11 Strata 3 Intensity 45 Threshold 10 63 Strata 4 Intensity 40 Threshold 9 Helpful Tips Ensure you can view the entire echogram if you can t see the range hash marks you need to expand the screen further with Toggle Header control and or manually dragging EG window Toggling between movie mode and echogram mode Right mouse click on echogram to see looped movie at same echogram time stamp Spacebar will pause loop Mark targets in EG mode with mouse left click and try to do so somewhere on target track because this also logs target range Right mouse click inside target marker to toggle to movie mode for manual measurement Zoom into target by mouse right click and diagonally dragging a focal window Use forward reverse arrow keys to advance through frames for optimum image measurement Note that direction of travel is logged based on how you measure the target head to tail Press lt F gt to log target Note that echogram will now indicate target length and direction of travel Blue marker downstream Yellow Upstream Some targets may hold linger in the beam and should not be tracked unless you deem them to have made sufficient progress up or down stream Significant over estimation of fish could occur if everything in the beam were tracked measured without regard fo
103. purpose Eon A T I T E SubNet SS EA Dell Latitude Laptop E6430 o Pa AS L Bank Offshore ARIS SubNet ARIS sonar BR T Dell Latitude Laptop E6400 J pe AS O O O O O Di O CEA EA R Bank Nearshore ARIS SubNet ARIS sonar N Dell Latitude Laptop E6430 ESA NetBooter o Z oo o o O oyoo Rightbankradio 128 95972 New NetBooter TS EZBridge Radio for ARIS A Left bank radio ssa J T 4 R Bank Offshore ARIS SubNet Dell Latitude Laptop E6430 128 95 97 150 admin Chinook fF o Radiolabs Bridge Radio for ARIS oo l S S Left bank radio 12895974 E E O S O Right bank radio 128 95 97 2 Hmmmm instructions say 1t uses 192 168 1 10 with sub net 255 255 255 0 A O O O O y yo R Bank Side Channel SubNet ARIS sonar a 1 V EZBridge Radio for ARIS o Right bank radio Backup Computers from 2012 A i t s ARII IM CIA E A AR laptop eer a ea V V AA AR 114 o Other info related to networks e The Didson networks for Right and Left Banks are NEVER connected together Therefore the separate laptops or the underwater units respectively can have the same IP addresses Of course no device can have the same IP address on the same network o The Local and Didson networks are not bridged together in any way 115
104. r Data is ENABLED Check Processing gt Show Parameters for correct direction of travel Motion for bank you are tracking Basic Tab and File Creation and Transmission Loss Advanced Tab Processing Parameters x Processing Parameters Basic 30 Advanced Transducer Basic 3D Advanced Transducer CSOT Cluster Echogram Image Transformation Process Angle fo Min Track Size 12 Factor A 0 999 Factor B 0 0009 Process N Beams fi y C Avg Over Threshold Factor C 0 65 Factor D 0 35 6 Max Over Threshold Convolve Beams 4 Convolve Samples 4 Min Cluster Area 300 Max Cluster rea 10000 Max Speckle 5 Max Fish Frame 32 Transmission Loss 2 Alpha R N Log R Alpha dB m M x Aut 0 3963 N 20 RAM Buffer 6 427 Mb pha dB m Man uto Eg C Normalize to 0 dB TL at Start Range Motion OS Y Show Cluster Statistics Normalize to 0 dB TL at End Range C Normalize to 0 dB TL at Range R R 1 C Upstream Motion R gt L Apply TL only to samples over Display Threshold Upstream Motion L gt R 6 Upstream Motion Subtract Opp Motion Limits File Creation Frame Range m Angle deg Pa EJ En 7 Auto Countiile N Bi File be Min 0 13 334 3 Jv Auto Countfile Name Biomass File td V Binary Count File dat V Echogram File ech M 4183 23 337 i Y New Countfile on Open Delete Empty Files ech V Entire File Jv All Ranges v All Beams Save Cancel Apply Defaults Save Cancel Ap
105. r Sockeye Main computer for on connection site data processing Right Bank Laptop NOT onboard 192 168 1 3 RM 8 6 dlburwen megaptera Sharing data from E but USB to Ethernet drive with Tactical Left Bank Laptop USB to 192 168 1 4 RM 8 6 dlburwen megaptera Sharing data from E Ethernet drive with Tactical PO O O O oo L Bank Offshore DIDSON SubNet DIDSON SONAR LR 128 95 97 227 RM8 6 381 e A AA a Ethernet rr Ferre LEE a DIDSON right bank SMC Bridge Radio for offshore 128 95 97 4 on RM 8 6 admin smcadmin DIDSON left bank L Bank Nearshore DIDSON SubNet Nearshore DIDSON laptop 2012 192 95 97 175 RM 8 6 dlburwen delphi 17 temporary computer used on both banks O e A a DIDSON right bank O e LEME fe DIDSON left bank R Bank Offshore DIDSON SubNet DIDSON SONAR LR 128 95 97 227 RM8 6 340 T S Right Bank DIDSON laptop onboard 128 95 97 203 RM 8 6 dlburwen megaptera Ethernet Nearshore DIDSON laptop 2012 192 95 97 175 RM 8 6 dlburwen Delphi 17 DFGANCDSF 164927 temporary computer R Bank Nearshore DIDSON SubNet Std DIDSON Anchor Rivers 128 95 97 227 RM8 6 2 Nearshore DIDSON laptop 2012 192 95 97 175 RM 8 6 dIburwen delphi 17 temporary computer used on both banks Appendix G2 Soldotna sonar office IP addresses SONAR Office IP address 112 S A Network Attached Network Attached Storage Western Digital Sharespace 4 GB ae 168 1 101 Sonar DU JS D oo SP R3 2011 data Wes
106. r if you have any questions regarding the information provided in this plan Regional Operational Plans are available on the Internet at http www adfg alaska gov sf publications Debby Burwen James Miller and Steve Fleischman Alaska Department of Fish and Game Sport Fish Division 333 Raspberry Road Anchorage AK 99518 This document should be cited as Burwen Debby L Miller James A and Steve Fleischman 2014 Kenai River Chinook Salmon Sonar Assessment Alaska Department of Fish and Game Regional Operational Plan ROP SF 2A 2013 23 Anchorage The Alaska Department of Fish and Game ADF amp G administers all programs and activities free from discrimination based on race color national origin age sex religion marital status pregnancy parenthood or disability The department administers all programs and activities in compliance with Title VI of the Civil Rights Act of 1964 Section 504 of the Rehabilitation Act of 1973 Title II of the Americans with Disabilities Act ADA of 1990 the Age Discrimination Act of 1975 and Title IX of the Education Amendments of 1972 If you believe you have been discriminated against in any program activity or facility please write ADF amp G ADA Coordinator P O Box 115526 Juneau AK 99811 5526 U S Fish and Wildlife Service 4401 N Fairfax Drive MS 2042 Arlington VA 22203 Office of Equal Opportunity U S Department of the Interior 1849 C Street NW MS 5230 Washington DC 20240 The d
107. r upstream progress Some targets may glance the corners of the beam if you feel these targets do not provide sufficient direction of travel information or that they may be dipping in an out of the beam throughout the sequence use your discretion at tracking When Chinook fall into this marginal category give them closer attention than smaller targets as missing a small target is less important than missing a KING Use all the tools at your disposal the echogram provides a roadmap to not only where targets occur in the sequence but often gives strong clues to what the target species is Trace intensity length and tail beat amplitude frequency give you information that comes in handy before you toggle to video mode Video mode and the video loop allow you to better sort out number behavior and location of targets in especially complex sequences Press lt E gt to save your work on each sequence when complete or before you divert to another task Hot Keys D Delete all line segments U Delete last line segment Space Bar Pause movie loop R L Arrow Advance Retreat movie frame L Click Drag Display movie loop of selected echogram time stamp R Click Drag Zoom into selected movie target area F Fish Target E Save tracked sequence 64 Example of Appropriate Target Picture and same with three segment manual measurement Known issues bugs Use of Fixed Background function Processing gt Background gt Fixed Background so
108. rate at high frequency at further ranges up to 30m than the standard model up to 15m The DIDSON LRs are equipped with an ultra high resolution lens to further improve image resolution A detailed discussion of available DIDSON configurations and DIDSON image resolution and a brief explanation of multibeam sonar can be found in Appendix A More detailed theory can be found in Belcher et al 2002 Electronics will be housed in a tent located on the right north bank of the river Figure 2 The DIDSONs will be mounted on remote pan and tilt systems a Remote Ocean Systems PT 25 on the right bank and a Sound Metrics Corporation X2 on the left bank for precise aiming in the horizontal and vertical axes The combined sonar and rotators will be deployed in the river on a tripod style mount Figure 3 In the horizontal plane the sonar will be aimed perpendicular to the flow of the river current to maximize the probability of insonifying migrating salmon from a lateral aspect Internal attitude sensors in the DIDSON will provide measurements of compass heading pitch and roll An AIM 2000 attitude sensor attached to the right bank mount will provide depth measurements throughout the season Communication cables from the right bank DIDSON feed directly into the right bank Top Side Box and data collection computer On the left bank DIDSON data is transmitted via a wireless bridge to a data collection computer on the right bank Figure 4 DATA ACQUI
109. rd Drive ARIS topside ARIS cable ARIS topside box Ethernet Wireless Bridge R Bank Radio 2 IP 128 95 97 4 Username new Password new Wireless Bridge L Bank Radio 1 IP 128 95 97 2 Username new Password new 2 TB External Hard Drive box USB 3 0 Left Bank Data Collection Computer Ethernet Right Bank Data Collection Computer 1 of 3 actual computers sonars 1 of 2 actual computers sonars On board Static IF none l static IF none DFGANCDSF50104093 Ethernet DFGANCDSF50104096 DFGANCDSF a0 104094 DFGANTCDSFS0 104095 Lisername sonar Password Count bish GlobalSat BU 353 DPGANCOSP50104097 Username sonar ees Password Count Fish Wireless Ethernet GlobalSat BU 353 AT amp T MiFi Liberate Hotspot VIF Network Name MiFI Liberate WEGA VIF Password 43387964 2013 phone 907 250 5280 Figure 20 ARIS data collection schematic for the RM 13 7 site For simplicity this diagram shows only one of three right bank data collection computer sonar pairs and one of two left bank data collection computer sonar pairs The wireless router will accommodate up to 8 computers 48 APPENDIX A DIDSON AND ARIS CONFIGURATIONS USED ON KENAT RIVER CHINOOK SONAR PROJECTS AT RM 8 6 AND RM 13 7 2013 49 Appendix A 1 DIDSON and ARIS configurations used on Kenai River Chinook sonar projects at RM 8 6 and RM 13 7 2013 including an overview of features that aff
110. reasonably good predictor of fish length Figure D1 2 Burwen et al 2010 the observed frequency distribution of Dlength supplies valuable information about species composition even though there is some overlap of Dlength measurements between species The Dlength mixture model is described below See also Fleischman and Burwen 2003 and Miller et al 2012 for a similar model using split beam echo length as the hydroacoustic variable The probability density function pdf of hydroacoustic variable y Dlength is modeled as a weighted mixture of two component distributions arising from sockeye and Chinook salmon Figure D1 3 fly nsfsly ne fe v D1 1 where fs y and fc y are the pdf s of the sockeye and Chinook component distributions and the weights Ts and Tc are the proportions of sockeye and Chinook salmon in the population Individual observations of y are modeled as normal random variates whose mean 1s a linear function of fish length x y Po t BX E D1 2 where Bo is the intercept B1 the slope and e is normally distributed with mean 0 and variance o Thus the component distributions fs y and fc y are functions of the length distributions fs x and fc x and the linear model parameters Bo P1 and o Figure D1 3 The species proportions ms and Tc are the parameters of Interest 73 Length measurements are obtained from fish captured by gillnets e g Eskelin 2010 immediately downstream of the sonar s
111. ributions from fitting this model provide prior distributions for DL mixture model Prior distributions in green font likelihoods in blue Age Mixture odc version 6a model Overall means and std deviations for a in 1 A sigma a dnorm 0 1 0E 4 I 0 taula lt 1 sigmaf a sigma a mula dnorm 0 1 0E 12 I 0 Dirichlet distributed age proportions across years within weeks D scale dunif 0 1 D sum lt 1 D scale D scale for win 1 W pi w 1 dbeta 0 2 0 4 pi 2p w dbeta 0 2 0 2 pilw 2 lt pi 2p w 1 pilw 1 pi w 3 lt 1 pi w 1 pi w 2 for y in 1 Y for a in 1 A D w y a lt D sum pi w a g w y a dgamma D w y a 1 A a g w y a sum g w y J for i in 1 nfish age i dcat pi wy week i year i 1 A length i dnorm mulageli tau ageli 83 Appendix D6 Example WinBUGS data under Fast Track sampling protocol SNR JULDATE 208 NET 202 lt JULDATE lt 208 NETTED FISH 295 TOTAL UPSTREAM 1038 NO THINNING list D species c 1 1 B1 2 4 B2 17 6 B3 3 5 B4 14 1 q1 a c 0 61 0 57 0 41 n fish 295 n meas 863 n small 175 speci s c 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 22 23222 DL a 22220 2 2220220272222 222222207220 70 D2 DDD DD Ds Dn DDD DD a DDD DD DS DAD DD A 22222220202 7 RSS PL OS DD JD DD ROSS ISSO Ils ES 151 222222202027 22D DI 2D ds DD DD DI JD OD DS DD 2222222020272 22D 2 2D ADA AI DD OD OD OA ISI AS 22222 I2 02027202
112. rnet Network adapter Hi Speed USB 23 orina aan GlobalSat BU 353 Username diburwen Password megaptera GlobalSat BU 353 Network Switch Ethernet Data Processing Computer Tactical Static IP for Shared Drive 192 168 1 2 On board ethernet Username SonarUser Password Sockeye Figure 6 DIDSON data collection schematic for the RM 8 6 sonar site 34 LB Data Storage Drive GO l OS 2012 05 1 49 Search 201 Organize v Include in library gt g e eSATAdirectto 4 TB Mi Computer 2 Name Left Bank Data External Hard Drive amp os c i J 2012 05 16_JD136 LB DIDSON Collection Computer 2012 05 16_000000_LHF ddf Y 2012 05 16_001000_LHF ddf 2012 05 16_002000_LHF ddf MZ 2012 05 16_003000_LHF ddf 2012 05 16_004000_LHF ddf MZ 2012 05 16_005000_LHF ddf WF 2012 05 16 010000_LHF ddf J 2012 05 16 JD137 J 2012 05 16_JD138 2012 05 16_JD139 J dell di Didson Data W Drivers eSATAdirectto 4 TB Right Bank Data i Ig External Hard Drive RB Data Storage Drive RB DIDSON Collection Computer GQ OS 2012 05 1 y 4 Search 201 P Organize v Include in library v a 0 j Computer 2 Name L os C J 2012 05 16_JD136 J 2012 05 16_JD137 J 2012 05 16_JD138 J 2012 05 16 JD139 J dell de Didson Data de Drivers Y 2012 05 16_000000_RHF ddf Y 2012 05 16_001000_RHF ddf Y 2012 05 16_002000_RHF ddf Y 2012 05
113. robability of a type I error rejection of the null hypothesis when true probability of a type II error acceptance of the null hypothesis when false second angular standard deviation standard error variance population sample Ha e CPUE CV F t X7 etc CI In log log etc NS Ho P Var var REGIONAL OPERATIONAL PLAN SF SA 2013 23 KENAI RIVER CHINOOK SALMON SONAR by Debby Burwen James Miller and Steve Fleischman Alaska Department of Fish and Game Sport Fish Division Anchorage Alaska Department of Fish and Game Division February 2014 The Regional Operational Plan Series was established in 2012 to archive and provide public access to operational plans for fisheries projects of the Divisions of Commercial Fisheries and Sport Fish as per joint divisional Operational Planning Policy Documents in this series are planning documents that may contain raw data preliminary data analyses and results and describe operational aspects of fisheries projects that may not actually be implemented All documents in this series are subject to a technical review process and receive varying degrees of regional divisional and biometric approval but do not generally receive editorial review Results from the implementation of the operational plan described in this series may be subsequently finalized and published in a different department reporting series or in the formal literature Please contact the autho
114. rom a dual frequency identification sonar DIDSON imaging system ICES Journal of Marine Sciences 63 543 555 Kucera Paul A 2009 Use of Dual Frequency Identification Sonar to determine adult Chinook salmon Oncorhynchus tshawytscha escapement in the Secesh River Idaho U S Department of Energy Bonneville Power Administration Environment Fish and Wildlife Portland OR Project 199702000 Contract 00035429 Maxwell S L and N E Gove 2007 Assessing a dual frequency identification sonar s fish counting accuracy precision and turbid river range capability Journal of the Acoustical Society of America 122 3364 3377 Miller J D D L Burwen and S J Fleischman 2010 Estimates of Chinook salmon passage in the Kenai River using split beam sonar 2006 Alaska Department of Fish and Game Fishery Data Series No 10 40 Anchorage http www sf adfg state ak us FedAidpdfs FDS10 40 pdf Miller J D D L Burwen and S J Fleischman 2013 Estimates of Chinook salmon passage in the Kenai River using split beam sonar 2008 2009 Alaska Department of Fish and Game Fishery Data Series No 12 73 Anchorage Mueller A T J Mulligan and P K Withler 2008 Classifying sonar images can a computer driven process identify eels North American Journal of Fisheries Management 28 1876 1886 Simmonds E J and D N MacLennan 2005 Fisheries Acoustics Theory and Practice Second Edition Blackwell Science Oxford W
115. s merely a quantitative version of this assessment in which the shape of the overall frequency distribution is modeled and fitted until it best approximates the data Uncertainty 1s assessed by providing a range of plausible species compositions that could have resulted in the observed frequency distribution As another example imagine that there are substantial numbers of small Chinook and that there 1s error in the length measurements The effect of the measurement error 1s to cause the modes to begin to overlap reducing the ability to detect detail in the length distribution and reducing the precision of the estimates Under this scenario it 1s still possible to make subjective assessments about the true species composition but to quantify the uncertainty is more difficult Mixture models provide an objective way to accomplish this Such a model can be conducted on any quantity related to length including length as measured from DIDSON images Dlength Given knowledge of the relationship between length and the observed quantity e g Burwen et al 2010 it 1s straightforward to convert from length units to the new units by including the slope intercept and mean squared error of the relationship in the mixture model Equation 10 below The more closely related the surrogate measurement is to the one of interest the more the two distributions will resemble each other and the better the resulting estimate will be Since Dlength is a
116. subsequently evaluated a second sonar site at RM 13 2 for using split beam sonar to assess fish passage but the bottom topography was less acoustically favorable and the fish were more difficult to detect due to increased background noise levels from bottom irregularities and boat traffic Burwen et al 2000 Because there were no other sites identified the idea of moving the site was abandoned The transition to DIDSON ARIS multibeam technology has reopened the option of moving the sonar program above tidal influence because the sonar beams are less sensitive to irregularities in the river bottom profile During 2012 a prospective site at RM 13 7 was identified Figure 1 and efforts to implement a full sonar site at this site will commence in 2013 In 2013 ADF amp G will run both the historical site at RM 8 6 and the second site at RM 13 7 One of the main advantages of the RM 13 7 site is the potential to achieve bank to bank coverage of the river with sonar which 1s not possible at the RM 8 6 site Like the sonar project at RM 8 6 the sonar project at RM 13 7 is scheduled to operate from May 16 August 10 2013 However because the project is considered experimental inseason estimates will not be produced in 2013 I OPERATIONS AT RIVER MILE 8 6 RM 9 RM 9 OBJECTIVES This portion of the study will provide real time estimates of the number of Chinook salmon passing river mile 8 6 to fishery managers during the 2013 fishing season Specific
117. surements panel g is modeled as a weighted mixture of species specific distributions b and e which in turn are the products of species specific size distributions a and d and the relationship between DIDSON measured length and true length c The weights species proportions panel f are the parameters of interest F4 Appendix D2 WinBUGS code for mixture model under standard data processing protocol Prior distributions in green font likelihoods in blue model beta0 dnorm 75 0 0025 subjective prior sd 20cm beta1 dnorm 1 25 subjective prior sd 0 2 sigma DL dunif 0 20 tau DL lt 1 sigma DL sigma DL ps 1 2 ddirch D species pa 1 1 dbeta B1 B2 theta1 dbeta B3 B4 pa 1 2 lt theta1 1 pa 1 1 pa 1 3 lt 1 pa 1 1 pa 1 2 pa 2 1 dbeta 0 5 0 5 theta2 dbeta 0 5 0 5 pa 2 2 lt theta2 1 pa 2 1 pa 2 3 lt 1 pa 2 1 pa 2 2 n chin lt ps 1 ntgts p large lt ps 1 1 pa 1 1 n large lt p large ntgts Lsig 1 1 lt 78 Lsig 1 2 lt 70 Lsig 1 3 lt 74 Lsig 2 1 lt 25 Lsig 2 2 lt 25 Lsig 2 3 lt 25 for s in 1 2 for a in 1 3 Ltau s a lt 1 Lsig s a Lsig s a mu 1 1 dnorm 621 0 0076 mu 1 2 dnorm 825 0 0021 mu 1 3 dnorm 1020 0 0047 mu 2 1 dnorm 380 0 0004 mu 2 2 dnorm 500 0 0004 mu 2 3 dnorm 580 0 0004 for ain 1 3 pa effective 1 a lt pa 1 a q1 a a inprod pa
118. surements at longer ranges resolved subsequent studies in 2008 2009 were directed toward developing an efficient process to generate DIDSON based estimates of Chinook salmon in a timely manner to meet the needs of fishery managers Each DIDSON LR system generates approximately 1 GB of video like data each hour and efficient ways of transferring reviewing and extracting size related data from such a large volume of data were needed A DIDSON LR HRL was deployed on the left bank of the Kenai River for 7 days in 2008 and 53 days in 2009 During this time efforts focused on developing and testing software for collecting and processing DIDSON data efficiently A relatively efficient procedure was developed to manually track and size individual fish Ways to automate the process of tracking and sizing individual fish were also explored during this time with limited success Miller et al 2012 In 2010 DIDSON was deployed on both banks to sample the same 60 m mid section of river insonified by the split beam sonar for most of the season 83 days 17 May 7 August To achieve a sufficiently high image resolution DIDSON sampled the river in 10 m range increments i e the shorter the range increment the higher the downrange resolution see by programming it to sequentially sample three 10 m range strata 3 13 m 13 23 m 23 33 m or 30m per bank Hardware focus problems related to changing range strata rendered substantial portions of 31 days of data u
119. t season Likewise a daily abundance model similar to that described on page 12 may be required post season to obtain daily estimates of Chinook salmon passage at RM 13 7 21 SCHEDULE AND DELIVERABLES Table 7 2013 timeline and milestones RM 8 6 site Normal camp set up with new computer DIDSONs Review update measurement protocol create a suite of test files that can be used with a tutorial for future training especially at site Group training on manual DIDSON measures for both site and office crews e Review update DIDSON based camp manual RM 13 7 site e Check with Mark Hatfield to ensure we have enough tripods and mounts 5 1 5 15 2013 e Construction of new tent platform new weatherport e Installation of ELP walkways IT needs mainly Mark s Duties e Update batch files for RM 8 6 DIDSON data continue with 4 range strata e Update SAS for RM 8 6 data processing and uploading e Develop new batch files for RM 13 7 ARIS data 5 10 Delivery date for 1 ARIS 1200 and 1 ARIS 1800 2 X2s and 2 LL 5 13 5 17 2013 Bill Hanot on site to assist with sonar testing and configuration e Deploy ARIS on RM 8 6 right bank when dust settles Conduct tethered fish experiments as early as possible to compare ARIS and DIDSON fish measurements e when netters start picking up at least 5 15 6 1 2013 several fish shift Develop field manual with examples and test files Training sessions with crew as needed for m
120. te Channel Template Channel 1 Length 775 Azmut 90 4 X 6043248 Aamutx 2697 X 6042360 Y 67105434 Y 67104837 Template Channel Template Channel Length 779 Azmut 2701 4 Amt 90 0 X 8043198 Y 6710548 Y 67104746 Template Channel Template Channel 4 Length 785 Azmut 90 2 X 6042205 6043224 8 Length 304 Azmutt 2704 X 60231 9 Y 67105199 Y 67104602 XY edtxyz Template Channel XY edtxyz Template Channel Figure 13 Corresponding profiles for nine transects conducted near river mile 13 7 of the Kenai River 41 RM 8 6 Site tent site 2012 Tent ad RM 13 75 site Site 6 010 Back channel LB downstream location Figure 14 Aerial view of sonar sites at Kenai river mile 8 6 top and 13 7 bottom with proposed sonar beam coverage Diagrams are approximate and not drawn to scale 42 Figure 15 Proposed coverage for the right bank top transect 5 in Figure 16 and left bank bottom transect 6 in Figure 16 at RM 13 7 43 Covers first 5 10m of ARIS 1200 range and 10m behind offshore sonar Covers first 5 10m of ARIS LR range and 10m behind offshore sonar Figure 16 Aerial view of sonar sites at Kenai river mile 8 6 top and 13 7 bottom with proposed sonar beam coverage Diagrams are approximate and not drawn to scale 44 lt ARIS cable Wireless Bridge R Bank Radio 2 IP 1284 95 97 4 Username new Pass
121. te the all segments lt space bar gt to pause in movie mode lt right arrow gt forward direction when you play movie or advances frame one at a time if the movie is pause d lt left arrow gt opposite of above Left Click Drag to show movie over the selected time Right Click Drag zooms the selected area 89 d Helpful Tips Toggling between BS mode and the raw image can sometimes be helpful in determining the actual end of the tail or snout wie Count Fish Some targets may hold linger in the beam and should not be tracked unless they make sufficient progress up or down stream Significant over estimation of fish could occur if everything in the beam were tracked measured without regard for upstream progress Some targets may glance the corners of the beam if you feel these targets do not provide sufficient direction of travel information or that they may be dipping in an out of the beam throughout the sequence use your discretion at tracking When Chinook fall into this marginal category give them closer attention than smaller targets as missing a small target is less important than missing a KING Use all the tools at your disposal the echogram provides a roadmap to not only where targets occur in the sequence but often gives strong clues to what the target species is Trace intensity length and tail beat amplitude frequency give you information that comes in handy before you toggle to video mode Video mode and th
122. tern Digital Sharespace 4 GB 192 168 1 103 Sonar Office Admin Chinook 2012 data Synology DiskStation 1512 10 192 168 1 165 Sonar Office Admin Chinook 2012 ARIS data TB backup Buffalo TeraStation Pro SA TB J a Computers _ Dell Workstation Laptop LAN 1 146 63 15 200 SOA network 146 63 5 200 SOA network domain login domain computer password Dell Optiplex 990 192 168 l xxx Sonar Office Jim s state Jim s state Backup data 146 63 15 xxx netwok domain login domain processing computer SOA network password Jim s Dell Optiplex 990 192 168 1 110 146 63 15 xxx Sonar Office sonargear smaLLfry2 ARIS DIDSON data SOA network processing Brandons DGSXQDCF 102790 146 63 15 xxx SOA network processing Kara s DGSXQDCF 102793 146 63 15 xxx SOA network processing Trevor s DGSXQDCF 102792 Dell Optiplex 9010 minitower J J J Nee Dell Optiplex 9010 minitower J J Nee SO Dell Optiplex 9010 minitower J J Nee SO Pe S V U DFGANCDS IB EN Sometimes in the office but now used as backups for sonar oe Trevor s former laptop field sites aia 16089 FISHSONAR Chinookl em s former laptop FISHSONAR Chinookl Brandon s former laptop Pinghappy2 Dell Latitude Laptop E6500 Dell Latitude Laptop E6500 Dell Latitude Laptop E6500 Map drives to local address 192 168 1 103 113 Appendix G3 RM 13 7 sonar site IP addresses 2013 configuration located
123. tes May 1 August 15 2013 Nathan Plate Fish and Wildlife Technician II Alaska Department of Fish and Game Assist Crew Leader with all aspects of DIDSON ARIS deployment operation and data processing Approximate work dates May 13 August 15 2013 Aaron Gordon Fish and Wildlife Technician II Alaska Department of Fish and Game Assist Crew Leader with all aspects of DIDSON ARIS deployment operation and data processing Approximate work dates May 13 August 15 2013 Alex Pettey Fish and Wildlife Technician II Alaska Department of Fish and Game 24 Duties Personnel Duties Personnel Duties Assist Crew Leader with all aspects of DIDSON ARIS deployment operation and data processing Approximate work dates May 13 August 15 2013 Cyndi Jaffa Fish and Wildlife Technician II Alaska Department of Fish and Game Assist Crew Leader with all aspects of DIDSON ARIS deployment operation and data processing Approximate work dates May 13 August 15 2013 Lindsay Fagrelius Fish and Wildlife Technician II Alaska Department of Fish and Game Assist Crew Leader with all aspects of DIDSON ARIS deployment operation and data processing Approximate work dates May 24 August 15 2013 25 BUDGETS The total proposed personnel requirements for the Kenai River Chinook Salmon Sonar Project at RM 8 6 fiscal year 2014 are summarized in Table 8 Table 8 Project personnel for RM 8 6 site
124. the formula Y W N Where W Window Length cm N number of range samples pixels With DIDSONs N is fixed at 512 samples pixels and images with shorter Window Lengths are always better resolved The DIDSON Window length parameter can only be set at discrete values 2 5 5 0 10 0 or 20 0 m for the DIDSON LR HRL at 1 2 MHz Although using shorter window lengths will increase resolution 1t will also require more individual stratum to cover the desired range Dividing the total range covered into too many discrete stratum increases the data processing time For this study a window length of 5m was used for the first two range strata to minimize the bias associated with close range targets see section d below A window length of 10 m was used for each of the two subsequent range strata sampled a compromise which allowed a relatively high resolution while allowing a reasonable distance to be covered by each stratum The down range resolution or pixel height for a 5m range window is 1 cm 500cm 512 and for a 10 m window length is 2 cm 1 000 cm 512 gt ARIS images can attain a finer down range resolution than DIDSON With ARIS N can vary to a maximum of 4 000 samples pixels and Window Length is user selectable This allows the user to collect data over a longer Window Length but increase the number of samples per beam to compensate Figure Al 2 contrasts images from a DIDSON LR HRL with an ARIS 1200 HRL The ARIS image in Figur
125. ttkktktktktik am Lx Fish 18 Range 8 47 Frame 5185 Bearing 99 0 Length 0 0 Thickness 0 0 More Del Undo Fish FA Pedvidd rr ad Oder 00 06 EIB AYORA RR OA IVA llep alas abd HA 2333 A A O E kl ick cr tra along ie from head to tal Hia wa Y lo add fa SES r x n Fish 18 Range 8 47 Frame 5185 Bearing 99 0 Length 95 1 Thickness 0 0 More Del Undo Fish For Help press FL AAA AAN Figure E1 2 On right ARIS images from a free swimming Chinook salmon showing the unzoomed image top the zoomed image middle and the segmented lines that result when the observer clicks along the length of the fish to mark its length bottom On left comparable DIDSON images from a tethered Chinook salmon The pixels of the ARIS image are less defined due to a smoothing algorithm applied Additionally the ARIS image has approximately twice the downrange resolution as the DIDSON image ARIS 20m 2000pixel Icm pixel resolution versus DIDSON 10m 512 2cm pixel 93 a 98 5 cm b 95 3 cm c 98 7 cm d 93 4 cm e 86 2 cm f 86 7 cm g 95 7 cm h 85 1 cm i 99 1cm dashed line or 88 1 cm solid line ee A T I 88 2 cm i 88 8 cm 95 5cm k 97 5cm 1 87 9 cm m 83 2 cm Figure E1 3 Panels a m show the potential variability in length measurements taken from images of a free swimming Chinook salmon at approximately 9 m collected with
126. ulative estimate of midriver Chinook salmon abundance and its variance is the sum across days Vig 24 0 v r v rl2 8 RM 9 MIDRIVER LARGE FISH PASSAGE ESTIMATES The daily estimate x of large fish passing RM 8 6 in mid river will be obtained with equations 1 4 after substituting C ij for Csij where C gjj number of upstream bound fish greater than 3 m from the right and left bank transducers exceeding 75 cm in length as measured by the DIDSON during tai NET APPORTIONED RM 9 MIDRIVER CHINOOK SALMON PASSAGE ESTIMATES The net apportioned daily estimate of midriver Chinook salmon abundance will be calculated by multiplying the RM 9 midriver salmon passage estimate by a netting derived estimate of the proportion of Chinook salmon Z ypz Perschbacher 2012 W y 7 NET 9 The variance estimate follows Goodman 1960 var 7 v r A er Phen v rls vae 4 xen v r 10 10 MODIFICATIONS UNDER FT PROTOCOL Under the Fast Track protocol length measurements will be available for all fish greater than 75 cm DL and a subset of fish 40 lt DL lt 75 Direction of travel will be available only for fish greater than 75 cm DL These constraints will require the following modifications to the abundance estimators Daily passage yrr will be defined as the number of salmon shaped fish gt 40 cm DL except downstream fish gt 75 cm DL It will be estimated as specified in equations 1 4 except that Cs
127. word new Wireless Bridge L Bank Radio 1 IP 1238 95 97 2 Username new Password new ARIS topside box j Left Bank Data Collection Computer Ethernet 1 of 3 actual computers sonars static IP none Right Bank Data Collection Computer 1 of 2 actual computers son rs Static IP none y On board Ethernet Username new Password new lt Wireless Ethernet AT amp T MiFi Liberate Figure 17 ARIS data collection schematic for the RM 13 7 site For simplicity this diagram shows only one of three right bank data collection computer sonar pairs and one of two left bank data collection computer sonar pairs The wireless router will accommodate multiple computers 45 The components shown inthe diagram below are housed inthe small white container inthe upper left corner of the fish tote 422928 The batteries are stored ina separate container inthe lower right corner of the fish tote The combined charger inverter are mounted inthe third container inthe lower left corner Ethernet or Coax to Bridze Topside Box 3007 ARIS cable CISCO Wireless Bridge Model AIR BR1310G A K9 R IPaddress 1285 95 97 X Subner Mask 255 255 255 0 Gateway XX XX XX XX XX XX 128 95 97 227 DIDSON IP Alnit MAC Address XX XX XX XX XK XX 7 Radio AMAC Address XX XXXX XX XXXX CISCO Wireless Bridge Model AIR BR1310G A KS R IP address 12895 97X Subnet Mask 255 255 255 0
128. y a number of batch files The names and functions of each batch file are listed in Individual files for each 10 minute sample will vary in size due to the different ping rates for each range strata File sizes will vary from a maximum of approximately 115 000 KB for the near range strata 3 7m and 7 13m 8 pings sec to approximately 101 000 KB for the mid range stratum 13 23m 7 pings sec and to approximately 78 000 KB for the offshore stratum 23 33m 6 pings sec At these data collections rates and ranges approximately 1 2 GB hour both banks or 30 GB per day worst case will be generated Since the season is about 87 days long up to 3 4 TB of data may be generated Data from each sample will be stored to a uniquely named file Filenames are automatically generated by the DIDSON software using year month day military time and frequency high or low For example a file that started data collection using high frequency on June 1 2013 at 3 40 AM will be automatically named 2013 06 01 034000 HF ddf One laptop will be dedicated to collecting data from each bank To ensure correct time stamps in the filenames laptop clocks will be synchronized using GlobalSat BU 353 Waterproof USB GPS receivers Data will initially be collected by the host computer hard drive and subsequently transferred to two 4 TB external hard drives two redundant copies for permanent archiving at the site Figure 6 Figure 7 Data collection computers will be n
129. y already be active as some of them have memory and are saved until changed 1 Select lt BS gt for background subtraction from toolbar or under lt Processing gt lt Background gt lt Background Subtraction gt Select lt Processing gt lt Background gt lt Fixed Background gt 3 Select threshold and range settings given in Table C1 1 To adjust these settings use the slider bars under Display Controls to the left of the echogram 4 Select the threshold and intensity settings for each range strata as indicated below To adjust these settings use slider bars under the Display Controls to the left side of the Echogram or Movie window 5 Select lt EG gt for view Echogram from toolbar or under lt Processing gt lt Echogram gt lt View echogram gt lt left click gt on the echogram near on the fish trace of interest to mark it you should see a white circle 7 lt right click gt INSIDE the white circle to switch to movie mode movie mode will play the 16 frames encompassing this circle continuously 8 Press lt space bar gt to pause movie Step through the movie frames using the right or left arrows until you find a frame that you think displays the entire length of the fish well see section C below for selecting optimal images 68 10 lt right mouse click drag gt will magnify the area in the rectangle 11 lt left click gt on the FISH SNOUT and continue to lt left click gt along the body to create a se

Download Pdf Manuals

Related Search

Related Contents

Téléchargez    RUMped by DL2RUM  Harbor Freight Tools Peak/700 Product manual  HIGH SPEED HDMI ™ DUAL CAT5e/6 EXTENDER WITH IR USER  en/ ACS355 drives quick installation guide  información de nuevo producto  BD SENSORS Aufsteckanzeige PA 430 - TRI  Graef WK 85 electrical kettle  Istruzioni per l`uso  

Copyright © All rights reserved. Failed to retrieve file