North American Journal of Fisheries Management ISSN: 0275-5947 (Print) 1548-8675 (Online) Journal homepage: http://www.tandfonline.com/loi/ujfm20 Precision of Three Structures for Saugeye Age Estimation Jeff Koch, Ben Neely & Bryan Sowards To cite this article: Jeff Koch, Ben Neely & Bryan Sowards (2017): Precision of Three Structures for Saugeye Age Estimation, North American Journal of Fisheries Management, DOI: 10.1080/02755947.2017.1394938 To link to this article: http://dx.doi.org/10.1080/02755947.2017.1394938 Accepted author version posted online: 25 Oct 2017. Submit your article to this journal View related articles View Crossmark data Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ujfm20 Download by: [University of Florida] Date: 26 October 2017, At: 09:50 Precision of three structures for saugeye age estimation Jeff Koch* and Ben Neely ip t Kansas Department of Wildlife, Parks, and Tourism, 1830 Merchant, Emporia, KS 66801, USA Bryan Sowards cr us *Corresponding author: firstname.lastname@example.org Abstract—We evaluated precision of age estimates obtained from sagittal otoliths, scales, and dorsal fin an spines from 341 saugeye (Walleye Sander vitreum × Sauger S. canadense) collected from six Kansas reservoirs. Between-reader agreement was greatest for otoliths, followed by dorsal spines, and then M scales. Coefficient of variation values for between-reader comparisons for otoliths were 1.3 to 3.1; whereas, those for dorsal spines were 13.8 to 18.2. Between-reader comparisons for scales were most ed variable, with CV estimates between 19.9 and 23.1. Age estimates from otoliths aligned with age estimates from scales more often than those from dorsal spines; although, age estimate comparisons ce pt between structures were generally variable. Between-reader agreement for scales and fin rays decreased with increasing estimated age; however, agreement for otoliths generally remained high (i.e., > 80%) through age 10. Given greater precision relative to other structures, we recommend the use of sectioned otoliths to estimate age of saugeye, especially when sacrificing fish is not a concern. Ac Downloaded by [University of Florida] at 09:50 26 October 2017 Kansas Department of Wildlife, Parks, and Tourism, #3 State Park Road, Sylvan Grove, KS 67481, USA Introduction An important consideration of age and growth investigations is precision of age assignments, which addresses repeatability of individual estimates, regardless of whether they are accurate ip t (Campana 2001). Although precision cannot be used as a surrogate for accuracy in the absence of true validation, precision is useful in assessing the relative ease of assigning age to a structure (Campana cr us comparing relative ages of structures obtained from the same fish. Biased age estimates can lead to erroneous estimates of population dynamics, harvest models, and population viability (Koch et al. 2009; an Hamel et al. 2016). As such, evaluating precision and the potential for bias in age estimates are important to fisheries managers. M Saugeye (Walleye Sander vitreum × Sauger S. canadense) are stocked by natural resource management agencies to provide angling opportunities and biological control of abundant fishes (e.g., ed centrarchids and Gizzard Shad Dorosoma cepedianum) in water bodies where Walleye and Sauger do not perform well (Mosher 2001; Boxrucker 2002; Galinat et al. 2002; Quist et al. 2010). Walleye x ce pt Sauger hybrids also naturally occur in systems with sympatric populations of parental species (e.g., Graeb et al. 2010). Although saugeye are commonly stocked by fisheries managers and are popular among anglers, there is a relative paucity of age and growth information for this hybrid, especially when compared to other recreationally important percids. A reason for this lack of information may be Ac Downloaded by [University of Florida] at 09:50 26 October 2017 2001). Additionally, researchers can identify relative biases along a gradient of estimated ages when difficulties associated with obtaining precise age estimates from saugeye hard structures (i.e., scales, dorsal spines, otoliths) or a lack of literature examining aging precision associated with these structures. Since saugeye are commonly used in attempt to balance predator-prey relationships, quality age and growth information is needed to properly manage saugeye populations and examine their influences on recreational fisheries. Several studies have investigated precision of structures used to estimate age of Walleye (Erickson 1983; Kocovsky and Carline 2000; Isermann et al. 2003), and have shown that otoliths generally provide more precise age estimates than scales or dorsal spines. Walleye otoliths have also been validated as an accurate hard structure for age determination to age 4 (Erickson 1983; Heidinger ip t and Clodfelter 1987; Spurgeon et al. 2015). Unlike Walleye, Sauger aging structures have not been validated and relatively few studies have compared precision of their age estimates. Dattilo et al. (2008) cr us the Missouri River. Additionally, Williamson and Dirnberger (2010) found age estimates obtained from whole and sectioned Sauger dorsal spines were similar. Saugeye and naturally occurring Walleye x an Sauger hybrids have generally been aged with whole or cracked sagittal otoliths (Denlinger et al. 2006; Graeb et al. 2010); although, Doyle (1999) attempted to age Oklahoma saugeye using scales but noted M difficulties in obtaining reliable age estimates with this method. To our knowledge, no studies have evaluated precision and comparative ages of hard structures used to age saugeye. The objective of this ed study was to estimate and compare precision of age estimates of saugeye obtained from sectioned sagittal otoliths, scales, and sectioned dorsal spines. ce pt Methods Saugeye were sampled in October 2016 using night electrofishing and gill nets at six Kansas reservoirs that varied in surface area from approximately 40 to 1,400 hectares. Study reservoirs and Ac Downloaded by [University of Florida] at 09:50 26 October 2017 indicated that otoliths provided more precise age estimates than scales or dorsal spines for Sauger in number of saugeye aged from each water included Madison City Lake (n = 9), Chase State Fishing Lake (n = 89), Geary State Fishing Lake (n = 94), Harvey County East Lake (n = 75), Marion County Lake (n = 23), and Kanopolis Reservoir (n = 51). Fish were measured for total length (TL; mm), euthanized, and transported to a laboratory for processing. Sagittal otoliths were extracted and stored in a plastic vial placed in a numbered scale envelope until further processing. Otoliths were later cleaned of residual tissue and mounted in epoxy following methods of Koch and Quist (2007). Encapsulated otoliths were sectioned transversely through the nucleus with a Buehler Isomet low-speed saw operated at 300 rpm (Buehler, Lake Bluff, IL). The third dorsal fin spine was removed as close to the body as possible with side cutters and placed in a scale envelope to dry. Dorsal spines were mounted in epoxy similar to ip t otoliths and sectioned as closely to the base of the structure as possible. Resulting otolith and dorsal fin spine sections were approximately 0.6 mm thick and viewed with the aid of immersion oil to improve cr us a numbered scale envelope to dry. Dried scales were cleaned of debris if necessary and aged without further processing. All structures were viewed with transmitted light on a stereoscope (10 – 40× an magnification) linked to a computer monitor. Structures were examined in their processed state so readers could manipulate the structure, lighting, or optics. M Three readers independently aged each structure without knowledge of the fish’s length. Readers 1 and 2 had several years of experience aging hard structures, including all three types used in ed this study. Reader 3 had minimal experience with all three structures but had previously examined scales and otoliths. When all three readers’ independent age estimates did not agree, a consensus age ce pt was agreed upon by all readers. Between-reader agreement of the same structure and agreement of each reader’s ages of different structures from the same fish were quantified as percent agreement (PA; when two age estimates were the same) and agreement within one year (PA1; when the absolute value of the difference between two age estimates was zero or one). Bias in age estimates was evaluated with Ac Downloaded by [University of Florida] at 09:50 26 October 2017 structure clarity. Scales were collected from the area directly posterior to the pectoral fin and placed in age bias plots using the ageBias function in R (Ogle 2015; R Core team 2016). This function uses onesample t-tests, which are corrected for multiple comparisons, to determine if mean age estimates for one age (i.e., x-axis) are significantly different from the average corresponding age estimates (i.e., yaxis). These t-tests essentially test for significant departures of mean age estimate values from the 1:1 line of equivalency. Stolarski and Sutton (2013) suggested that bias is evident when deviations from the line of equivalence occur for more than two successive years. Precision of age estimates (i.e., among readers and structures) was also evaluated using the coefficient of variation (CV; Chang 1982; Campana ip t et al. 1995). cr us number of times each fish is aged. an Results Sectioned otoliths, scales, and sectioned dorsal spines were collected from 341 saugeye varying M in TL from 130 mm to 670 mm (Figure 1). Between-reader agreement was generally greatest for otoliths, followed by dorsal spines and scales (Figure 2). Agreement between readers varied from 96.2% ed to 98.0% for otoliths and CV values for reader combinations were between 1.3 and 3.1. No bias between readers was detected for otoliths for any age estimate combination. Agreement between ce pt readers for scale age estimates varied from 64.1% to 68.8% and CV values ranged from 19.9 to 23.1. Significant deviations between readers occurred at young ages (i.e., ages 0 and 1); however, patterns in bias thereafter were not evident in older age classes. Between-reader agreement varied from 70.5% to 77.7% for sectioned dorsal spines and CV values were between 13.8 and 18.2. Significant deviations Ac Downloaded by [University of Florida] at 09:50 26 October 2017 where Xij is the ith age determination of the jth fish, Xj is the mean age of the jth fish, and R is the from equivalence were evident between readers from ages 0 to 4. The three readers agreed on otolith ages 95.6% of the time with a mean CV of 3.2. No consistent bias was indicated by greater than two consecutive years between readers for any structure. Most deviations from equivalency between readers occurred at estimated ages 4 and less. Dorsal spine age estimate agreement with otolith age varied by reader from 55.7% to 63.6% and coefficient of variation values for the comparison between the two structures varied from 17.0 to 21.7 (Figure 3). When the age estimates for the two structures differed for presumed younger fish, dorsal spine age estimates were typically greater than otolith ages; however, for older saugeye, significant ip t deviation from the line of equivalency were evident with otolith ages being greater than those from dorsal spines. The relationship between otolith and scale age estimates was the least variable structure cr us relation to other comparisons, varying from 63.5% to 80.7% agreement. When dorsal spine ages were compared to scale ages, agreement varied by reader from 49.7% to 70.0%. Eight of nine age bias plots an illustrated bias between structures that occurred for at least two consecutive age estimates; however, there were never three consecutive years of significant bias in any plots. Thus, persistent bias was not M observed. In most comparisons for fish estimated to be 0 or 1 yr by otoliths, scales and dorsal spines had significantly greater ages compared to otoliths. Furthermore, for fish estimated as age 0 and age 1 ed by dorsal spines, scales generally had greater estimates as well. In general, agreement of age estimates was greater for saugeye estimated to be less than three ce pt years old (Figure 4), and agreement generally decreased for purportedly older saugeye. Percent agreement among all three readers for structures with consensus ages greater than three years was 79.1% for otoliths, 16.6% for scales, and 45.8% for dorsal fin spines. For saugeye with a consensus otolith age greater than three years, between-reader agreement varied from 83.3% to 87.5%. For Ac Downloaded by [University of Florida] at 09:50 26 October 2017 comparison in regard to CV values, and the two structures also generally had the highest agreement in saugeye with consensus scale ages greater than three years, between-reader agreement rate was 20.8% to 29.2%; however, this value varied from 54.1% to 66.7% for similar comparisons of sectioned dorsal spines. Discussion Biologists must make several decisions when choosing an aging structure, including whether sacrificing fish is acceptable, removal and processing times of structures, and precision and accuracy of ip t resulting age estimates. Our results indicated that age estimates obtained from sectioned otoliths exhibited no bias between readers and were more precise than those obtained from sectioned dorsal cr us spines, especially at young ages, indicating inconsistencies in identifying annuli from early ages. Although between-reader agreement was greater for scales and dorsal spines at young ages, bias was an more consistent at young ages, leading to significant deviations from equivalency. Our between-reader agreement rates were slightly greater than those reported by Isermann et al. (2003) who evaluated M Walleye scales, sectioned dorsal spines, and sectioned otoliths; however, mean CV values were similar. Our results were also similar in regard to between-reader agreement reported for scales, sectioned ed dorsal spines, and cracked otoliths from Sauger collected from the Missouri River (Dattilo et al. 2008). Although CVs from between-reader comparisons were four to ten times greater for scales and dorsal ce pt spines, collection of these structures is not lethal and requires considerably less removal and processing time compared to otoliths (Isermann et al. 2003). Although Isermann et al. (2003) indicated that wholeview otolith age estimates were the most time-efficient and precise approach for estimating age of walleye, cursory examinations of whole saugeye otoliths used in our study did not reveal readily Ac Downloaded by [University of Florida] at 09:50 26 October 2017 spines and scales. Conversely, significant bias was indicated between readers for scales and dorsal observable annuli. As such, we used sectioned otoliths, whose age estimates were highly precise and relatively easy to read compared to dorsal spines and scales. Although our study used sectioned otoliths, our results were similar to those of Isermann et al. (2003), who reported that scale ages more closely reflected whole-view otolith ages compared to sectioned dorsal spines. Isermann et al. (2003) asserted that inconsistencies and difficulties associated with precisely aging sectioned Walleye dorsal spines were associated with identifying the first annulus. We also noted difficulties identifying the inner annulus on dorsal spine sections, and likely led to inconsistencies in age estimates between readers. This bias between readers was evident for dorsal spines at young ages as significant deviations from the line of equivalence was noted in every reader ip t comparison for at least one year for saugeye estimated to be age three or younger. A consideration for future examinations of saugeye dorsal spines is the findings of Williamson and Dirnberger (2010), who cr us illuminated Sauger dorsal spines, the side-illumination technique was more time efficient and displayed outer annuli on older fish better than sectioning. This method may be useful for future examinations of an saugeye dorsal spines as dorsal spine ages were significantly less than those obtained from otoliths in purportedly older fish. Futhermore, Logsdon (2007) suggested that side-illuminated Walleye dorsal younger Walleyes. M spines provided a reasonably precise and non-lethal approach for replicating otolith ages for age-7 or ed Although sectioned otoliths yielded precise age estimates in our study, we were unable to assess the accuracy of estimates. Furthermore, no published study has validated the use of any hard ce pt structure for use in estimating age of saugeye. As such, validation studies of saugeye age estimates are warranted. Although we cannot assess accuracy of age estimates of saugeye in our study, age estimates obtained from walleye otoliths have been validated as accurate to age 4 (Erickson 1983; Heidinger and Clodfelter 1987; Spurgeon et al. 2015). As such, until age estimates of saugeye are validated, biologists Ac Downloaded by [University of Florida] at 09:50 26 October 2017 indicated that although reader agreement and precision were similar for sectioned and whole, side- must consider options regarding the choice of saugeye hard structure. If fish sacrifice is not acceptable, scales more closely represented otolith ages (i.e., as evidence by higher percent agreement and lower variability between structures) compared to sectioned dorsal spines; however, sectioned dorsal spines yielded more precise age estimates. Regardless, we recommend sectioned otoliths for age estimation of saugeye if fish sacrifice is not a concern. Acknowledgements We thank Paul Stockebrand for preparing aging structures. We also thank John Reinke, Craig Johnson, Sean Lynott, Tyler Thomsen, Chris Steffen, and Paul Stockebrand for assistance with sampling. This ip t paper was improved by comments from Martin Hamel, Michael Quist, and four anonymous reviewers. Funding was provided by Kansas Department of Wildlife, Parks, and Tourism as well as Sport Fish cr us References an Boxrucker, J. 2002. Improved growth of a White Crappie population following stocking of saugeyes (Sauger × Walleye): A top-down density-dependent growth response. North American Journal M of Fisheries Management 22:1425-1437. Campana, S. E. 2001. Accuracy, precision, and quality control in age determination, including a review ed of the use and abuse of age validation methods. Journal of Fish Biology 59:197-424. ce pt Campana, S. E., M. C. Annand, and J. I. McMillan. 1995. Graphical and statistical methods for determining the consistency of age determinations. Transactions of the American Fisheries Society 134:866–871. Chang, W. Y. B. 1982. A statistical method for evaluating the reproducibility of age determination. Ac Downloaded by [University of Florida] at 09:50 26 October 2017 Restoration Grant F-22-R. Canadian Journal of Fisheries and Aquatic Sciences 39:1208-1210. Dattilo, J. E., R. R. Dirnberger, P. T. Horner, D. J. Niswonger, M. L. Miller, and V. T. Travnichek. 2008. Three year summary age and growth report for Sauger Sander canadensis. Missouri Department of Conservation, Chillicothe. Denlinger, J. C. S., R. S. Hale, and R. A. Stein. 2006. Seasonal consumptive demand and prey use by stocked saugeyes in Ohio reservoirs. Transactions of the American Fisheries Society 135:12-27. Doyle, W. J. 1999. Impacts of saugeye introductions on growth and population structure of Largemouth ip t Bass and White Crappie in a turbid reservoir. Master’s thesis, Oklahoma State University, Stillwater. cr Erickson, C. M. 1983. Age determination of Manitoban Walleyes using otoliths, dorsal spines, and us Galinat, G. F., D. W. Willis, B. G. Blackwell, and M. J. Hubers. 2002. Influence of a saugeye (Sauger × an Walleye) introduction program on the Black Crappie population in Richmond Lake, South Dakota. North American Journal of Fisheries Management 22:1416-1424. M Graeb, B. D. S., D. W. Willis, N. Billington, R. N. Koigi, and J. A. VanDeHey. 2010. Age-structured assessment of Walleyes, Saugers, and naturally produced hybrids in three Missouri River ed reservoirs. North American Journal of Fisheries Management 30:887-897. ce pt Hamel, M. J., J. J. Spurgeon, C. J. Chizinski, K. D. Steffensen, and M. A. Pegg. 2016. Variability in age estimation results in ambiguity and false understanding of population persistence. North American Journal of Fisheries Management 36:514-522. Heidinger, R. C., and K. Clodfelter. 1987. Validity of the otolith for determining age and growth of Ac Downloaded by [University of Florida] at 09:50 26 October 2017 scales. North American Journal of Fisheries Management 3:176-181. walleye, striped bass, and smallmouth bass in power cooling plant ponds. Pages 241–251 in R. C. Summerfelt and G. E. Hall, editors. Age and growth of fish. Iowa State University Press, Ames. Isermann, D. A., J. R. Meerbeek, G. D. Scholten, and D. W. Willis. 2003. Evaluation of three different structures used for Walleye age estimation with emphasis on removal and processing times. North American Journal of Fisheries Management 23:625-631. Koch, J. D. and M. C. Quist. 2007. A technique for preparing fin rays and spines for age and growth analysis. North American Journal of Fisheries Management 27:782-784. Koch, J. D., M. C. Quist, C. L. Pierce, K. A. Hansen, and M. J. Steuck. 2009. Effects of commercial harvest ip t on Shovelnose Sturgeon populations in the upper Mississippi River. North American Journal of Fisheries Management 29:84-100. cr Kocovsky, P. M. and R. F. Carline. 2000. A comparison of methods for estimation ages of unexploited us Logsdon, D. E. 2007. Use of unsectioned dorsal spines for estimating Walleye ages. North American an Journal of Fisheries Management 27:1112-1118. Mosher, T. 2001. An evaluation of saugeye (Stizostedion vitreum × S. canadense) in Midwestern waters M and their effect on panfish populations: a study by the Walleye Technical Committee of the North Central Division of the American Fisheries Society. Kansas Department of Wildlife, Parks, ed and Tourism, Pratt. ce pt Ogle, D. H. 2015. Introductory fisheries analyses with R. CRC Press, Boca Raton, FL. R Core Team. 2016. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Spurgeon, J. J., M. J. Hamel, K. L. Pope, and M. A. Pegg. 2015. The global status of freshwater fish age Ac Downloaded by [University of Florida] at 09:50 26 October 2017 walleyes. North American Journal of Fisheries Management 20:1044-1048. validation studies and a prioritization framework for further research. Reviews in Fisheries Science and Aquaculture 23:329-345. Stolarski, J. T. and T. M. Sutton. 2013. Precision analysis of three aging structures for amphidromous Dolly Varden from Alaskan arctic rivers. North American Journal of Fisheries Management 22:732-740. Quist, M. C., J. L. Stephen, S. T. Lynott, J. M. Goeckler, and R. D. Schultz. 2010. An evaluation of angler harvest of Walleye and saugeye in a Kansas Reservoir. Journal of Freshwater Ecology 25:1-7. Williamson, C. W. and R. R. Dirnberger. 2010. A comparison of techniques using dorsal spines to cr us an M ed ce pt Ac Downloaded by [University of Florida] at 09:50 26 October 2017 ip t estimate Sauger age. North American Journal of Fisheries Management 30:1016-1019. Figure 1. Length distribution of saugeye aged using sectioned otoliths, scales, and sectioned dorsal cr us an M ed ce pt Ac Downloaded by [University of Florida] at 09:50 26 October 2017 ip t spines. Saugeye were sampled from six Kansas reservoirs in October 2016. Figure 2. Age-bias plot with estimates of mean coefficient of variation (CV), between-reader percent agreement (PA), and percent agreement within one year (PA1) for three readers of sectioned sagittal otoliths, scales, and sectioned dorsal spines from saugeye collected from Kansas reservoirs, 2016. Error bars represent 95% confidence intervals around the mean age assigned by one reader relative to fish ip t assigned an age by a second reader. Open circles represent significant deviation from the line of cr us an M ed ce pt Ac Downloaded by [University of Florida] at 09:50 26 October 2017 equivalency. Figure 3. Age-bias plots with estimates of mean coefficient of variation (CV), between-reader percent agreement (PA), and percent agreement within one year (PA1) for three readers of sectioned sagittal otoliths, scales, and sectioned dorsal spines from saugeye collected from Kansas reservoirs, 2016. Error bars represent 95% confidence intervals around the mean age assigned by one reader relative to fish ip t assigned an age by a second reader. Open circles represent significant deviation from the line of cr us an M ed ce pt Ac Downloaded by [University of Florida] at 09:50 26 October 2017 equivalency. Figure 4. Between-reader agreement of three readers (i.e., R1, R2, R3) by consensus age of sectioned sagittal otoliths, scales, and sectioned dorsal spines from saugeye collected from Kansas reservoirs, cr us an M ed ce pt Ac Downloaded by [University of Florida] at 09:50 26 October 2017 ip t 2016.