SC/A06/HW53 19th Century Southern Hemisphere Humpback Whale Catches Tim D. Smith Cachalot Project, Woods Hole, MA, USA Elizabeth Josephson Northeast Fisheries Science Center, Woods Hole, MA, USA Randall R. Reeves Okapi Wildlife Associates, Hudson, Quebec, Canada
Abstract 19th century open-boat whalers killed humpback whales in all ocean basins except the Antarctic, with the majority from the southern hemisphere. Townsend’s (1935) charts of the location of whaling vessels when humpbacks were taken show low latitude catches especially along the west coasts of Africa and South America, around some oceanic islands such as Tonga, and along the east and west coasts of Australia. Best (1987) used data from Townsend’s (1935) whaling logbook study to estimate the total global 19th century catches. Recent digitization of roughly half of Townsend’s original data sheets allows estimation of the proportion of humpbacks taken in different regions. We used Best’s (1987) estimates of total global catches with Townsend’s data to estimate catches for several regions, which in some cases represent putative humpback populations. Catch data extracted directly from a small sample of logbooks confirmed the general pattern of catches seen in Townsend data, and sightings data from the logbooks revealed a more extensive distribution pattern than shown by catches alone. The data summarized here allow assignment of the global catches to regions, and these regional totals need to be compared to subsequent catches to determine their significance. Introduction 19th century open-boat whalers took humpback whales in all ocean basins except the Antarctic. Best (1987) combined Townsend’s (1935) logbook-derived tabulations of baleen whale catches (16,837 whales taken in 1,651 voyages) with production data from Starbuck (1878) and Hegarty (1959) to estimate “landed catches” of bowhead, right, humpback, and gray whales, by five-year period, between 1805-1909. He estimated that roughly 14,000 to 18,000 humpbacks were killed and processed by the American offshore fleet in the 19th century. We digitized approximately half of Townsend’s original data sheets, and we also extracted catch, sightings and other data from a small sample of logbooks of voyages not included in Townsend’s study. The three datasets (Best’s, Townsend’s and our own) were used to describe the extent of humpback whaling in several oceanic regions, including twelve areas thought to represent southern hemisphere populations or stocks (SC/57/SH11).
1
Materials and Methods We obtained copies of Townsend’s (1935) original data sheets for voyages by vessels with names beginning with the letters A-J. The data were transcribed into computer files. The digitized dataset included the dates and catch positions of 1174 humpback whales recorded in 108 logbooks between 1790 and 1910. Of the 1174 catches, 1164 took place between 1800 and 1900. Townsend reported locations by either latitude and longitude or place name. In cases of the latter, we referred to navigation charts to assign approximate coordinates. The proportions of the 1164 catches that occurred in fourteen oceanic areas were computed, including ten regions defined by the IWC Scientific Committee as having distinct populations or stocks of southern hemisphere humpbacks (SC/57/SH11). Numbers of humpbacks taken in each region were estimated as the product of those proportions times the midpoint of Best’s (1987) two alternative estimates (one based on oil returns and average yield, the other on average number of humpbacks secured per voyage) of total humpback removals – 16,188. Daily vessel locations along with whale sighting and catch data were extracted from over 100 complete logbooks, 75 of which included humpback whale catches and 15 of which were for the same voyages used by Townsend (1935). The locations of sightings and catches of humpback whales were plotted. Results The distribution of humpback whales in the available data was similar to that shown on Townsend’s (1935) chart (compare Figures 1 and 2), with a substantial proportion of the catch positions in the southern hemisphere. The proportions of whales caught in the 14 regions varied substantially (Table 1). The numbers of whales taken by decade varied markedly, with most between 1840 and 1890 (Figure 3). Estimates of the numbers taken by decade are given in Table 2. Humpback whales reported in our sample of logbooks as sighted but not killed were distributed more widely than those that were killed (Figure 4). Discussion Townsend’s (1935) chart demonstrated that sizeable catches of humpback whales were made by American 19th century whalers in the southern hemisphere. His chart was somewhat difficult to interpret quantitatively, however, because it showed only the noon positions of whale ships on days when one or more humpback whales were taken. Our digitized sample of Townsend’s original data allowed the proportions in different areas and over time to be computed. Most humpbacking in the 19th century occurred between 1840 and 1890, and most of that was in regions where southern hemisphere populations occur (Figure 2). Based on Best’s (1987) estimate of global humpback catches, the numbers that may have been caught from each of the ten southern hemisphere populations ranged from zero to over 4000, with most taken in the regions of Western Africa, Western South America, and Tonga (Table 1, Table 2). Estimates of catches on the West Indies and Cape Verde Islands breeding grounds from 1865 to 1886 have been made: 1617 (SE 278) and 441 (SE 126), respectively (Smith and
2
Reeves 2003). These values can be compared to our much cruder estimates from the period 1860 to 1890 (Table 2) from the eastern and western North Atlantic because Townsend’s catch data in the North Atlantic were primarily from these two breeding areas (Figure 1). The sum of the estimates for the eastern North Atlantic, 375, is very similar to that for the Cape Verde Islands and well within statistical uncertainty of Smith and Reeves’ (2003) estimate alone. The sum for the western North Atlantic is substantially lower, but probably within statistical uncertainty given were that for Best’s (1987) estimate known. The general consistency between Smith and Reeves (2003) and our present estimates gives some support for the admittedly crude method we are using here for other regions. The reliability of the estimates presented in the present paper depend critically Best’s (1987) estimates, which in turn depend on assumptions about the representativeness of Townsend’s logbook sample, the accuracy of the numbers of voyages and oil returns reported by Starbuck (1878) and Hegarty (1959), and the average oil yield of humpbacks (Best used 24.4 (CV=0.110) barrels/whale). These new estimates also depend on the representativeness of our digitized sample of Townsend’s data. Best (1987) rightly characterized his estimates as first approximations and acknowledged a number of uncertainties. Without knowing the statistical uncertainty in Best’s (1987) total estimates and in the proportions that we have used to generate the present regional estimates, the overall statistical uncertainty can not be determined. Regarding some of the foundational assumptions of Best’s (1987) analysis, we have identified several concerns. Lund (2002) identified additional voyages not included in Starbuck (1878) or Hegarty (1959). Further, comparison of Lund’s list of voyages and Townsend’s logbook sample reveals that he under sampled voyages by smaller vessels between 1860 and 1890. Similarly, comparison of the dates of catches recorded by Townsend in the available abstracts and the departure and return dates of the corresponding voyages from Lund (2001) suggests that some of the logs Townsend used may have been incomplete. The effects of these uncertainties on the estimates given here, and indeed on Best’s original estimated totals, needs to be further examined. Finally, it is well known that a substantial proportion of humpback whales that were struck were not actually secured and processed, and so do not show up in the estimates given here. As pointed out by Best (1987), correction factors to account for this requires the examination of logbooks. A multiplier of 1.85 was used for the recent North Atlantic humpback assessment (IWC 2003), and our preliminary examination of logbooks from the Gulf of Guinea suggest that loss rates on at least one Southern Hemisphere humpbacking ground were as high as on the North Atlantic breeding grounds. The 1.85 correction factor incorporates consideration of orphaned and injured calves. The data from the logbooks also indicated that calves were frequently killed; the calf component of total removals would likely not be accounted for in Best’s (1987) estimates. The data summarized here allow assignment of the global catches to regions, and these regional totals need to be compared to subsequent catches to determine their significance. Total removals undoubtedly exceeded the landed catches by a considerable extent, and it would be important to also account for animals struck but lost and for orphaned calves in a complete assessment of the effects of 19th century whaling in the Southern Hemisphere.
3
References Best, P.B. 1987. Estimates of the landed catch of right (and other whalebone) whales in the American fishery, 1805-1909. Fish. Bull. 85:403-18. Hegarty, R.H. 1959. Returns of Whaling Vessels Sailing from American Ports. A Continuation of Alexander Starbuck’s ‘History of the American Whale Fishery’ 18761928. Old Dartmouth Historical Society, and Whaling Museum, New Bedford, Massachusetts. 58pp. Lund, J. 2001. Whaling Masters and Whaling Voyages. Sharon, Massachusetts: Kendall Whaling Museum. Starbuck, A. 1878. History of the American Whale Fishery from its Earliest Inception to the Year 1876. Report of the U.S. Fish Commission 4, 1875-6, Appendix A. Smith, T.D. and R.R. Reeves. 2003. Estimating American 19th Century Catches of Humpback Whales in the West Indies and Cape Verde Islands. Carrib. J. Sci. 39:286-297. Townsend, C.H. 1935. The distribution of certain whales as shown by logbook records of American whaleships. Zoologica (N.Y.) 19(1-2):1-50 + 6 maps.
Figure Legends Figure 1. Numbers of humpback whales caught, by 5º square, in our sample of Townsend’s (1935) logbooks. Figure 2. Electronic scan of Townsend’s (1935) chart of daily ship locations when one or more humpback (and bowhead) whales were reported caught. Figure 3. Number of humpback whales in our sample of Townsend’s (1935) logbook data that were from southern hemisphere and northern hemisphere populations, by decade. Figure 4. Location of humpback whales sighted (open circles) and caught (closed circles) in our sample of logbooks, showing that the catches were localized in only part of the range over which whales were sighted.
4
Table 1. Number and proportion of humpbacks in Townsend’s (1935) sample taken in the 19th century in each of ten regions containing southern hemisphere populations (SC/57/SH11) and four northern hemisphere half ocean basins (Regions), with estimated total number of whales caught. Region Number Proportion Est. Totals A: E. South America 15 0.013 209 B: W Africa 291 0.250 4047 C: E. South Africa 54 0.046 751 D: W. Australia 93 0.080 1293 Ei: E. Australia 66 0.057 918 Eii(1): New Caledonia 3 0.003 42 Eii(2): Tonga 204 0.175 2837 F: Cook Islands 0 0 0 G: W South America 260 0.223 3616 X: Arabian Sea 0 0 0 E North Atlantic 40 0.034 556 W North Atlantic 30 0.026 417 E North Pacific 81 0.070 1126 W North Pacific 27 0.023 375 Totals 1164 1.000 16188
5
Table 2. Estimates of humpbacks caught from several regions (see Table 1) and in total, by decade of voyage departure,. Region 1800s 1810s 1820S 1830s 1840s 1850s 1860s 1870s 1880s 1890s
A B C D Ei Eii(1) Eii(2) F G X W NAtl E N Atl W N Pac E N Pac Total
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 28 0 0 0 0 0 28
0 0 0 0 0 0 0 0 0 0 14 0 0 0 14
0 14 681 0 0 0 0 0 0 0 0 0 0 0 695
28 14 56 111 0 0 0 0 56 0 0 14 14 14 307
0 403 14 932 14 0 0 0 375 0 42 14 139 70 2003
6
181 640 0 250 695 28 0 0 681 0 264 42 960 292 4033
0 2350 0 0 209 14 2837 0 250 0 139 111 14 0 5923
0 528 0 0 0 0 0 0 2225 0 97 223 0 0 3073
0 97 0 0 0 0 0 0 0 0 0 0 0 0 97
1900s 0 0 0 0 0 0 0 0 0 0 0 14 0 0 14
1-3 4-50
Figure 1. Numbers of humpback whales caught, by 5º square, in our sample of Townsend's (1935) logbooks.
51-100 101-231
400 200 0
100
Whales in Sample
300
S. Hemisphere N. Hemisphere
1800
1820
1840
1860
1880
Decade
Figure 3. Number of humpback whales in our sample of Townsend’s (1935) logbook data that were from southern hemisphere and northern hemisphere populations, by decade.
! (
! ( ! ( ! (
! ( ! (
! ( ! ( ! (
! (! ! ( ( ! ( ! (
! ( ! ( ! ( ! (
! ( ! ( ! ( ! ( ! ( ! ! ( ( ! (! ! ( ( ! ( ! ( ! (
! ( ! ( ! (
Figure. 4 Location of humpback whales sighted (open circles) and caught (closed circles) in our sample of logbooks, showing that the catches were localized in only part of the range over which whales were sighted.
! (
