Sonar Sampling Protocol and Data Generation
From 1981-1994, pelagic fish abundance data were collected along a set of transects in each of six lakes (Allequash, Big Muskellunge, Crystal, Sparkling, Trout, Mendota) using a Simrad 70 khz EY-M echosounder. The transducer was attached to a 4 aluminum towbody suspended in front of the boat and deployed at a speed of approximately 3-4 knots. Transects were run on two nights and two days in late summer in each year such that they intersected the deepest portions of each lake. The returning acoustic signal was recorded on audio tape (until ~ 1989) or DAT tapes (from ~ 1989-1994), as well as on paper charts. The recorded signal was analyzed with the deconvolution program developed by C.S. Clay (Rudstam et al. 1987, Stanton and Clay 1986, Jacobson et al. 1990) and with the HADAS post processing package by Torfinn Lindem (Lindem 1993, Rudstam et al. 1988) to estimate fish densities, by size, for each discrete depth in the lake. Most of the information collected from 1981 to 1989 was collected without recording the gain setting on the audio tapes and some tapes were recorded with too low gain resulting in too high signal to noise ratios. This made post processing difficult, however some data could be recovered by using the target strength of the dominant fish species to scale the recordings. Lars Rudstam analyzed data prior to 1989 using target strength estimated from fish caught in gillnets to calibrate the sonar information. Data and information was published on Trout Lake for 1983 and 1985 (Jacobson et al. 1990), Trout and Muskellunge Lakes for 1981 (Rudstam et al. 1987), on Mendota for 1981 to 1989 (Rudstam et al. 1993) and expanded to 1991 in DeStasio et al. (1995). For Crystal Lake, Rudstam generated data from 1981-1988 while Hrabik analyzed information from 1989-1995 (Sanderson et al. 1999). In 1995, the Simrad EY-M echosounder ceased to work reliably. In 1996, the LTER project purchased an HTI Model 241 echosounder with a 120 kHz split beam configuration. This echosounder was deployed in the manner described above on (Allequash, Big Muskellunge, Crystal, Sparkling, Trout, Mendota, Monona, and Fish Lakes). Ecoscape post-processing software, produced by HTI, was used to post-process data. Data were archived in the output format from HTI sounder software v. 1.0 and raw acoustic signals were stored on digital audio tapes. Prior to post processing of all HTI data, however, the computer containing the Ecoscape software ceased to work. No computer was purchased to replace it and the analysis, in 1998-9, also ceased. However, all the raw acoustic information is archived on digital audio tapes and processed on the HTI sounder software output files. After the laptop that operated the HTI system failed, there were no funds offered to replace it. No information was collected in 2000 because there was no laptop.
Thus, there have been two major changes in analysis methods over time. The first was a change in single beam methods from the C.S. Clays deconvolution method to T. Lindems HADAS system. Rudstam et al (1988) found the two methods comparable. The second change involved switching from single beam analysis to split beam, from a 70kHz frequency to 120kHz and from Simrad to HTI and later Biosonics. Rudstam et al. (1999a) compared the single beam HADAS analysis using 70kHz (Simrad EYorM, HADAS analysis), split beam 70kHz (Simrad EY500, EP500 analysis) and split beam 120 kHz (Simrad EY500, EP500 analysis) for rainbow smelt in Lake Erie. Differences in density estimates and average target strengths were not large although there was a bias in the HADAS approach to single beam derived average target strength of 0.8dB (Rudstam et al. 1999a). Rudstam et al (1999b) reviewed the single beam methods in general and Mason and Schaner (2001) has compared data from the Biosonics, Simrad, and HTI units for smelt in Lake Champlain.
From 2001-2003, sonar data was collected on Trout, Sparkling and Crystal Lakes using a Biosonics DT-6000 Echosounder with a 120kHz split beam transducer (T. Hrabik). Post-processing was performed using Echoview (SonarData Inc.) analysis software. In 2004, a Biosonics DT-X echosounder with a 70 kHz split beam transducer was used on Trout and Sparkling Lakes (T. Hrabik). No information was collected on Crystal Lake (the generator made too much noise in 2003 and caused a response from Law Enforcement). The information collected by Hrabik between 2001 and 2004 is currently being analyzed to generate aggregated lake-wide and 200 m transect-level fish size and density estimates (which can be converted into biomass and biomass by species using gillnet information) as well as transect-level data stratified at a 1m vertical depth resolution.

Data Correction:

December 2013: an error was detected for data from Sparkling Lake for the year 2004. Erroneous data have been replaced with corrected data in the metadata version 10 and data version v3.

January 2014: an error was detected for data from Trout Lake for the year 2012. Erroneous data have been replaced with corrected data in the medatadata version 11 and data version v4.

Literature cited
DeStasio, B. J., L. G. Rudstam, A. Haning, P. Soranno, and Y. Allen. 1995. An in situ test of the effects of food quality on Daphnia population growth. Hydrobiologia 307:221-230.
Jacobson, P. T., C.S. Clay, and J.J. Magnuson. 1990. Size, distribution, and abundance of pelagic fish by deconvolution of single beam acoustic data. Rapp. P.-v. Reun. Cons. int. Explor. Mer 189:304-311.
Lindem, T. 1983. Successes with conventional in situ determination of fish target strength. FAO Fish. Rep. 300:104-111.
Lindem, T. 1990. Hydro acoustic data acquisition system HADAS. Instruction manual. Lindem data acquisition Lda, Oslo.
Lindem, T., and D. A. Houari. 1988. Hydro acoustic data acquisition system HADAS. mimoegraphed report.
Lindeman, R. L. 1942. The trophic dynamic aspect of ecology. Ecology 23:157-176.
Mason, D. M., and T. Schaner. 2001. Final report to the Great Lakes Fisheries Commisison for the acoustics intercalibration exercise in 1999.
Rudstam, L. G., C. S. Clay, and J. J. Magnuson. 1987. Density and size estimates of cisco, Coregonus artedii using analysis of echo peak a single transducer sonar. Canadian Journal of Fisheries and Aquatic Sciences 44:811-821.
Rudstam, L. G., S. Hansson, T. Lindem, and D. W. Einhouse. 1999. Comparison of target strength distributions and fish densities obtained with split and single beam echo sounders. Fisheries Research 42:207-214.
Rudstam, L. G., T. Lindem, and S. Hansson. 1988. Density and in situ target strength of herring and sprat: a comparison between two methods of analyzing single beam sonar data. Fisheries Research 6:305-315.
Rudstam, L. G., T. Lindem, and G. LaBar. 1999. The single beam analysis. Pages 6-13 in E. Ona, editor. Methodology for target strength measurements (with special reference to in situ techniques for fish and micronekton). International Council for the Exploration of the Sea, Copenhagen.
Sanderson, B. L., T. R. Hrabik, et al. 1999. Cyclic dynamics of a yellow perch (Perca flavescens) population in an oligotrophic lake: evidence for the role of intraspecific interactions. Canadian Journal of Fisheries and Aquatic Sciences 56: 1534-42.
Stanton, T. K., and C. S. Clay. 1986. Sonar echo statistics as a remote-sensing tool: volume and seafloor. IEEE Journal of Oceanic Engineering OE-11:79-96.