Status of the Lake Trout Fishery in Lake Superior

The production of lake trout in the United States waters of Lake Superior was low (only 1,465,000 pounds) in 1879, the first year for which there is a record. Expansion of the fishery must have started soon thereafter, for the take was 3,488,000 pounds in 1885, the next year for which we have statistics, and averaged 3,416,000 pounds in 1885–1892. The years after 1892 can be divided readily into three general periods with average yields as follows: 1893–1907–4,599,000 pounds; 1908–1925–2,168,000 pounds; 1926–1949–3,049,000 pounds. A take of 3 million pounds can be held as “normal” in the modern fishery. During the three periods just listed the percentage contributions of the individual states to the United States total (as computed from the averages for individual periods) ranged from 65.1 to 71.5 for Michigan, 17.5 to 25.6 for Wisconsin, and from 9.3 to 11.0 for Minnesota. In the Canadian (Province of Ontario) waters of Lake Superior the average annual output of lake trout rose from 309,000 pounds in 1871–1882 to 900,000 pounds in 1883–1893, 1,567,000 pounds in 1894–1903, and 2,189,000 pounds in 1904–1918. This last period of relatively high yield was followed by two intervals of successively lower average catches–1,691,000 pounds in 1919–1929 and 1,395,000 pounds in 1930–1949. For the combined United States and Canadian waters of Lake Superior the general trends in the production of lake trout can be described by the following averages: 1879–1,653,000 pounds: 1885–1892–4,325,000 pounds; 1893–1907–6,236,000 pounds; 1908–1949–4,403,000 pounds. The short‐term fluctuations of production during the more recent years give evidence of periodicity in the output of lake trout in Michigan, Ontario, and in the entire lake. Furthermore, these periodic fluctuations tended to be similar in Michigan and Ontario waters. The coefficient of correlation (r) between production in Michigan and Ontario in 1920–1949 (after elimination of trend in the statistics for both areas) had the significant value of 0.456. This correlation suggests that Michigan and Ontario fishermen exploit a common stock or stocks subject to similar fluctuations. Statistics on the production of lake trout in 5 of the 6 statistical districts of the State of Michigan waters of Lake Superior (see Fig. 2 for boundaries of the districts) in 1885 suggest that in most areas the fishery was then in the process of development. Even in 1891–1908 when the general level of production was high, there is evidence that during certain periods the catch in some areas was influenced strongly by factors (such as accessibility to market) other than the natural productivity of the waters. Comparisons of the average annual output of lake trout in the individual districts in 1891–1908 and 1929–1943 (the base period for our modern statistical analysis) reveal an enormous decrease in the Whitefish Bay region (S‐6) from 916,000 pounds in 1891–1908 to only 177,000 pounds in 1929–1943, a substantial drop (from 655,000 pounds to 385,000 pounds) in the Marquette‐Munising area (S‐4), and a small decrease (from 141,000 pounds to 138,000 pounds) in the Black River‐Ontonagon district (S‐2). Among the remaining districts the average yearly take increased from 322,000 pounds in 1891–1908 to 354,000 pounds in 1929–1943 at Isle Royal (S‐1), from 428,000 pounds to 501,000 pounds in the Grand Marais district (S‐5), and from 422,000 pounds to 506,000 pounds in the Keweenaw area (S‐3). These changes in the catch resulted in a westward shifting of production centers. Districts S‐1, S‐2, and S‐3 which together contributed only 30.7 percent of the 1891–1908 catch accounted for 48.4 percent of the take in 1929–1943. The 1929–1949 production in all districts exhibited periodic fluctuations that were characterized by peaks in the middle 1930ˈs and middle 1940ˈs. The indices of abundance or availability as computed from records of the catch of lake trout per unit of fishing effort showed periodic fluctuations in all districts of the State of Michigan waters similar to those of production (the peaks and the intervening minima fell a little earlier in the curves of abundance than in the production curves). For the combined districts the abundance of lake trout, expressed as a percentage of the 1929–1943 mean, stood at 108 in 1929, dropped to 100 in 1931, rose to a 21‐year high of 137 in 1934, decreased to 80 in 1940, increased again to 107 in 1944 and then fell to the 21‐year low of 65 in 1949 (this last decline was interrupted by a small increase in 1947). The 1929–1949 fluctuations in abundance were similar in the 5 mainland districts (S‐2 through S‐6). The coefficients of correlation (r) were significant for all 10 pairings and those for the 4 easterly districts (S‐3 through S‐6) were extremely high (p < 0.001 for all 6 values). Thus we have evidence that the fishermen along the mainland exploit common stocks or stocks in which the factors controlling availability are the same or subject to closely similar fluctuations. The fluctuations in abundance in S‐1 were correlated significantly (p < 0.05) with those in S‐2 but otherwise appeared to be independent of conditions along the mainland. The distinctly cyclic fluctuations that characterized the statistics on production and abundance were much less apparent in the 1929–1949 data on fishing intensity in the State of Michigan districts. In districts S‐3, S‐4, and S‐5 a possible tendency toward a cyclic fluctuation seems to have been obscured by a long‐term upward trend that was becoming stronger toward the end of the 21‐year period. For the combined districts the level of fishing pressure was consistently high after 1943. Over the 6‐year period, 1944–1949, fishing intensity expressed as a percentage of the 1929–1943 mean averaged 142; for the most recent 4 years, 1946–1949 the average intensity index was 151. On the whole, the 1929–1949 production of lake trout in the State of Michigan waters of Lake Superior was unreliable as an indicator of changes in abundance. The coefficient of correlation between the fluctuations of catch and abundance was, to be sure, positive and significant in S‐1 (p < 0.05) and S‐6 (p < 0.01), but in the remaining districts and for the combined districts the values of the coefficient were far below the level of significance. The failure of production to serve better as an indicator of changes in abundance can be attributed to the negative correlation that existed between abundance and fishing intensity (values of r highly significant in every district but S‐6 and for the combined districts). The relationship suggests that fishermen have increased their fishing pressure in order to maintain their production during the recent years of declining abundance. The condition of the lake trout fishery of the State of Michigan waters of Lake Superior must be termed unhealthy and the outlook for the future is not good. Production in 1949 stood at 106 percent of the 1929–1943 mean, but this level of yield was made possible only by fishing intensity that was 162 percent of average; the abundance index in 1949 was only 65. Certainly the stocks of lake trout are in a poor state to withstand the threatened inroads of sea lampreys which have been taken from all parts of Lake Superior and are known to have established spawning runs at least as far west as the Keweenaw Peninsula.