![]() ![]() For investigation of chemical makeup related to source area, the data were analyzed via ordination, which positions sites (or specimens) along axes of variation (gradients) drawn through multidimensional space. Only four elements (Sr, Na, Mg and Ba) were consistently noted in suf fi cient abundance to be used in multivariate analysis. #Plash island history fullThe full data set is available upon request to the second author. Including standards, blanks, and in- dividual raster lines on otoliths, 305 sample datasets were obtained. Level of detection was set at 3 standard deviations (s.d.) above background. Frequency was 5 Hz, pulse was 45 mJ, and argon fl ow rate was 1.5 L min À 1. Given that the fi sh were relatively young (Reitz et al., 2013), and therefore fast-growing, the initial (outer) raster line should produce data representing chemical inputs from the waters where the fi sh were captured. This correspondence is approximate, however, as magni fi cation was too high to allow exact matching of raster lines and micromill channels to be made, a problem that may be correctable in future comparative work by staining of milled specimens or shading via directed light. The laser raster was set to 500 m m-long lines running parallel with the micromilled lines. ![]() Line interval was set at 50 m m, with the number of lines set to correspond with the micromilled part of the otolith previously excavated for light isotope analysis (Reitz et al., 2013). For our study, the equipment was calibrated daily and sample order alternated between sites. Still other sources of analytical noise are the buildup of ablated material in the machine (e.g., Ca accumulation on the skimmer cones e Campana et al., 1994), which necessitates frequent cleaning, and isotopic interference of 40 Ca with 40 Ar in the carrier gas (e.g., Barnett-Johnson et al., 2005), which limits the choice of analytes. (Interestingly, the effect varied by suites of elements in fi sh from different water bodies e Chang et al., 2012). (2012) compared 10, 20, and 30 m m beam widths and, not surprisingly, obtained the best results with the largest width, which they estimated represented 30 e 45 days of growth for hake ( Merluccius merluccius ). be below detection limits (e.g., Chang et al., 2012), so there is tradeoff between temporal precision and data acquisition. ![]()
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |