Catch statistics for belugas in West Greenland 1862 to 1999

M.P. Heide-Jørgensen1 and A. Rosing-Asvid2

1 Greenland Institute of Natural Resources, c/o National Marine Mammal Laboratory, 7600 Sand Point Way NE, Seattle WA98103, USA 2 Greenland Institute of Natural Resources, Box 570, DK-3900 Nuuk, Greenland

ABSTRACT

Information and statistics including trade statistics on catches of white whales or belugas (Delphinapterus leucas) in West Greenland since 1862 are presented. The period before 1952 was dominated by large catches south of 66o N that peaked with 1,380 reported kills in 1922. Catch lev- els in the past five decades are evaluated on the basis of official catch statistics, trade in mattak (whale skin), sampling of jaws and reports from local residents and other observers. Options are given for corrections of catch statistics based upon auxiliary statistics on trade of mattak, catches in previous decades for areas without reporting and on likely levels of loss rates in different hunting operations. The fractions of the reported catches that are caused by ice entrapments of whales are estimated. During 1954-1999 total reported catches ranged from 216 to 1,874 and they peaked around 1970. Correcting for underreporting and killed-but-lost whales increases the catch reports by 42% on average for 1954-1998. If the whales killed in ice entrapments are removed then the cor- rected catch estimate is on average 28% larger than the reported catches.

Heide-Jørgensen, M.P. and Rosing-Asvid, A. 2002. Catch statistics for belugas in West Greenland 1862 to 1999. NAMMCO Sci. Publ. 4:127-142.

INTRODUCTION factor in the total catch of belugas) was prohib- ited in 1995.

unting of white whales or belugas The catch statistics for belugas go back to 1862 (Delphinapterus leucas) in West but are not complete and have never incorporat- HGreenland has been the subject of con- ed animals that are killed but lost. Especially troversy because of indications that the catches for the most recent decades, the quality of the exceed sustainable levels (Heide-Jørgensen and catch statistics has fluctuated because of a large Reeves 1996, Butterworth et al. 2002, Heide- number of catches that are not included in the Jørgensen and Acquarone 2002, Innes and statistics. This fraction of unreported catches Stewart 2002). Several regulations have been and the number of killed-but-lost whales cannot introduced during the 1990s with the intention be estimated precisely, but estimates can be of reducing the catches. Since 17 October 1995 made under certain assumptions. The catch his- the catch of belugas from larger vessels has tory of belugas in West Greenland has, together been restricted, so that vessels from 25 to 50 with abundance estimates, recently been used Gross Register Tons (GRT) are only allowed to for modeling the dynamics of this stock and for catch belugas for their own consumption, not predictions of sustainability of future harvest for sale. Vessels from 50 to 79.9 GRT are only levels (Butterworth et al. 2002, Innes and allowed to take 2 belugas per year. Further- Stewart 2002, Alvarez-Flores and Heide- more, the drive hunt (which was a significant Jørgensen MS 2000). For the modeling of the

127 NAMMCO Scientific Publications, Volume 4 population trajectory it is essential that the actu- catches that are taken from whale pods that are al removals caused by humans be estimated as entrapped in the ice. It has been suggested that accurately as possible. Erroneous removal lev- mortality in ice entrapments occasionally is part els will result in unrealistic predictions of the of the natural mortality (Siegstad and Heide- productivity of the stock. Jørgensen 1994). To allow for analyses of re- movals without catches in ice entrapments In this compilation the uncorrected catch statis- these are shown separately from the mortality tics are presented together with corrections pre- genuinely caused by humans. viously applied (e.g. Kapel 1983, Born and Kapel 1986, Born 1987). There has been sys- MATERIAL AND METHODS tematic underreporting caused either by an ab- sence of catch reports from some municipalities Available statistics during certain periods or by a general trend to- The main sources for data on beluga catches in wards underreporting. In an attempt to correct Greenland are the official catch statistics. for these biases two levels of corrected catches Statistics on purchasing of mattak (whale skin) are introduced for the period after 1954 (low can under certain assumptions be converted to and medium options). Finally, a third level (high numbers of whales killed. The number of lower option) corrects the catches for an estimate of jaws collected from the harvest provides a min- killed-but-lost whales in different catching op- imum estimate of the number of whales that erations. The three levels are intended as op- were killed. tions for modeling different scenarios of the dy- namics of the population. Catch statistics before 1993 This compilation utilises statistics on beluga Catch statistics for belugas and narwhals catches in West Greenland between 1862 and (Monodon monoceros) from Greenland include 1891 from Winge (1902) and Anon. (1944), be- tween 1907 and 1951 from Heide- Jørgensen (1994) and Anon. (1944), be- tween 1954 and 1974 from Kapel (1977), between 1975 and 1990 from unpub- lished statistics from the Ministry of Greenland, Kapel (1983), Kapel and Larsen (1984), Kapel (1985), Born and Kapel (1986), Born (1987) and Heide- Jørgensen (1994). For the periods 1862- 1921, 1934-1948 and 1954-1963 catches were reported for the period 1 April through 31 March, but for the tabulation here all catches were allocated to the first of the years reported.

Catch statistics after 1992 A booklet called ‘Piniarneq’ in which the hunters note their catches has since 1993 provided catch figures. It operates with a hunting season from 1 October to 30 September but data are compiled follow- Fig. 1. ing calendar years. Map of West Greenland with important towns, Purchasing of mattak settlements and Data on purchasing of mattak is available localities indi- from Heide-Jørgensen (1994), Upernavik cated. Museum (in litt.) and the Greenland Sta-

128 Belugas in the North Atlantic and the Russian Arctic tistics (in litt.). From the three northernmost com- Corrections for underreporting after 1954 munities Uummannaq, Upernavik and Qaanaaq Known catches that are not included in the re- (see Fig. 1) most mattak is purchased by factory ports or catches that can be derived from statis- facilities. The average mattak yield per whale tics on trade in whale products were added to was estimated as 70 kg for belugas in Uper- the catch reports to improve the completeness navik based on measurements in the field of the reports. (Heide-Jørgensen 1994). To correct for catches in areas and periods with- Collection of jaws out reporting where catches are known to have Lower jaws have been collected from the catch- occurred, two levels of correction factors (low es of belugas as part of the harvest monitoring and medium options) were applied. The two op- program by the Greenland Institute of Natural tions are intended to correct for systematic neg- Resources since 1992. ative bias in the reporting.

Allocation of catches to areas For 1954-1974 a low and a medium option cor- For the period up to 1951 the catches from the rect for lack of catch reports from Qaanaaq. southernmost areas, Qaqortoq, Paamiut, Nuuk During the 21 years with reporting there were and Maniitsoq, are combined under ‘South of only 8 years with data from Qaanaaq and of 66° N’. It is suspected that a separate stock these 6 years were reported to be incomplete found in this area was extirpated through ex- (Table 1). The average annual catch of belugas ploitation (Heide-Jørgensen 1994). For the peri- over the 8 years was 63 (95% CI: 18-109). od after 1954 the catches are split into three ar- eas of West Greenland. The northern area For 1975-1985, corrections were based on the includes Qaanaaq (formerly Avanersuaq but reported trade in mattak. The correlation be- changed to Qaanaaq in 1997), Upernavik and tween the number of whales caught and the Uummannaq. The central area or Disko Bay in- amount of mattak purchased in the 1980s cludes Qeqertarsuaq, Ilulissat, Qasigiannguit, showed that approximately 65% of the mattak Aasiaat and Kangaatsiaq. The southern area in- harvested was sold to factories (Heide- cludes Sisimiut, Maniitsoq, Nuuk, Paamiut and Jørgensen 1994). From 1981 through 1985, re- Qaqortoq. ports exist of the mattak yield in Upernavik (see Table 2), almost all of it from belugas (Heide- Allocation of catches by species Jørgensen 1994). If the mattak yield represents Catches of narwhals and belugas were com- 65% of the catches of belugas (Table 1) then it bined in the annual statistics before 1906 can be calculated that 550, 466, 464, 455 and (Winge 1902, Heide-Jørgensen 1994) and in the 128 belugas were caught during 1981-1985 in statistics on average catches before 1939 Upernavik. Reported catch was 81% (Table 1) (Anon. 1944). For the years with combined sta- of this estimated catch on average in Uper- tistics, catches from Sisimiut and south are all navik, so this was chosen as a correction factor considered to be belugas because of the rare oc- for reported catches in Upernavik (low option) currence of narwhals in this area (Winge 1902, and in all areas (medium option). Heide-Jørgensen 1994). For the areas north of Sisimiut, Winge (1902) stated that the catch For 1986-88, corrections of total catches were was predominantly belugas and Anon. (1944) provided in the official statistics, but since they claimed that, based on trade in narwhal tusks, were not applied to certain areas it seems more narwhals only constituted about 20% of the appropriate to make corrections for specific catch north of Sisimiut. When comparing belu- municipalities. In the official statistics from the ga and narwhal catches for 1954-1970 (Anon. Greenland Home Rule, bulk corrections of 75, 1953-1971) the proportion of belugas taken 90 and 275 are applied for 1986, 1987 and 1988 north of Sisimiut was about 75%. That value is respectively. These numbers were used to cor- applied for the proportion of belugas taken rect all areas and they are not included in this north of Sisimiut during 1862-1905 and for av- compilation. For the same reason a block esti- erage catches reported during 1906-1934. mate of 750 belugas taken in 1992 in all munic-

129 NAMMCO Scientific Publications, Volume 4 ENTRAPMENT 503 309 50 1,874 1,774 pel (1983), Kapel and 98. The year 1999 only 98. eq’. ing or ‘ALL’ if it is a ge- d Qeqertarsuaq. The catches be- d Qeqertarsuaq. NAAQ NAVIK NAQ QAQORTOQ REPORTING IN ICE . Catches of belugas from official reports by municipality with corrections for under-reporting (in parentheses) for 1954 to 19 reports by municipality with corrections for under-reporting . Catches of belugas from official able 1 covers the period from January through September. The column ‘under-reporting’ covers the period from January through September. shows the sum of corrections for under-report neral correction factor for all areas. ‘Disko Bay’Aasiaat, Qasigiannguit, Ilulissat an includes the municipalities Kangaatsiaq, fore 1975 are extracted from Kapel (1977), between and 1990 unpublished statistics the Ministry of Greenland, Ka (1994), and from 1993 to 1999 ‘Piniarn Larsen (1984), Kapel (1985), Born and (1986), (1987) Heide-Jørgensen 1954195519561957195819591960196119621963 16 321964 10 851965 9 751966 61 125 61967 150 3 31968 121969 8 13 15 1,774 111970 91971 4 12 18 2751972 4 6 17 6 201973 373 3911974 23 2 7 121975 25 182 243 6 53 11 34 (47) 50 179 21 219 97 29 111 334 95 88 186 93 66 238 35 42 166 214 65 1 293 36 17 47 150 6 262 5 195 398 23 3 369 8 1,013 1 25 24 19 8 33 1,133 661 15 1 24 328 8 76 12 362 46 18 268 100 581 4 512 10 11 1 13 123 11 47 135 38 11 130 121 40 135 9 24 18 14 12 4 11 4 4 30 8 1 41 300 14 424 33 25 70 225 345 1 216 329 229 2 488 331 561 596 1,259 978 (47) 1,524 739 841 234 654 50 1,050 913 1,067 T YEAR QAA- UPER- UUMMAN- DISKO BAY SISIMIUT MANIITSOQ NUUK PAAMIUT- UNDER- TOTAL MORTALITY

130 Belugas in the North Atlantic and the Russian Arctic ENTRAPMENT (550)(687) 550 687 (ALL: 25) 275 125 NAAQ NAVIK NAQ QAQORTOQ REPORTING IN ICE Continued. able 1. 19761977 (37) 501978 (36) 5019791980 201981 251982 77 301983 240 761984 (10) 53 1271985 1041986 12 49 21 2501987 190 (165) 233 1911988 343 44 329 953 22 (60) 333 (135) 188 379 100 19 95 17 (335) 500 (100) 194 452 15 379 6 72 412 43 (50) 99 550 (150) 352 340 4 313 125 (75) 177 77 35 (25) 25 109 13 13 (25) 25 2 62 114 95 5 12 16 45 29 65 48 17 23 125 1 13 17 18 1 1 8 78 2 8 (36) (37) (325) 1,212 839 601 6 (235) 653 (425) 763 719 611 741 888 1,017 220 (335+ALL 75) 894 695 100 (ALL 90) 696 198919901991 (2) 2 (50) 50 (346) 346 (400) 400 427 (311) 8 2 (591) 684 (100) 100 (18) 30 23 40 (939) 1,063 (339) 499 500 1992199319941995 1191996 241997 841998 (169) 339 (661) 6611999 7 16 (90)188 51 194 (111) 26 21 19 18 86 162 (26) 26 194 162 42 239 21 28 301 38 80 14 105 245 243 116 312 25 116 38 131 101 56 125 14 30 26 7 10 3 19 1 4 25 1 2 18 30 6 1 (169) 1 7 798 (111) (90) 6 780 618 542 576 744 239 T YEAR QAA- UPER- UUMMAN- DISKO BAY SISIMIUT MANIITSOQ NUUK PAAMIUT- UNDER- TOTAL MORTALITY

131 NAMMCO Scientific Publications, Volume 4 ipalities (Heide-Jørgensen 1994) is ignored. No those that sink might be retrieved either by official statistics are available for 1991 and hooks in the following days or when floating to 1992. the surface. The losses during the drive fishery are therefore considered minimal and a catch For 1986-1992, two corrections are needed to correction factor of 1.10 has been applied to the include catches from Qaanaaq and Sisimiut. reported catches to correct for whales lost dur- The mean of reported catches during 1978 to ing drive fishery operations. The drive fishery 1985 (Table 1) was 68 (95% CI: 16 to 119) in was the most important way of hunting belugas Qaanaaq and 70 (95% CI 46 to 94) in Sisimiut. in Qaanaaq and Upernavik until it was banned The low and the medium option are the addition in 1995 (effective from the 1996 hunting sea- of the lower and upper bounds, respectively, of son). the confidence interval to the reported catches in Qaanaaq (1990-91 excepted) and Sisimiut. Shooting whales in open or ice-covered waters (=non-drive fishery) usually takes place over Correction for losses deep water and wounded whales are in many The bulk of the catches of belugas in West cases not retrieved. The non-drive fishery there- Greenland are taken either in a drive fishery or fore has a much larger proportion of lost whales by shooting in open water, along ice edges or in and a catch correction figure of 1.30 (see ice cracks. There are also a limited number of Discussion) was applied to the statistics from net catches. During the drive fishery, the whales this type of hunting. This hunt type is practiced are shot in shallow water after being driven to in all areas south of Upernavik and from 1996 the shore. Few whales escape the drives and even in Upernavik and Qaanaaq.

Table 2. Purchases (in tons) of beluga mattak per area (data from 1965-1992 from Heide- Jørgensen (1994), Upernavik 1975-1978 from Upernavik Museum and from 1994 to 1998 com- piled by Greenland Statistics).

Year Ilulissat Qeqertarsuaq Uummannaq Upernavik Qaanaaq Total 1965 1.51.5 1966 0.15.35.4 1967 0.13.94.0

1975 2.92.9 1976 0.41111.4 1977 9.11423.1 1978 2.868.8 1979 14 14 1980 14 14 1981 25.026.551.5 1982 21.224.045.2 1983 21.112.833.9 1984 20.721.942.6 1985 5.85.8 1986 1.71.7 1987 0 0 4.432.6?37 1988 0 4.97.327.32.441.9 1989 0 0.831.019.3?51.1 1990 16.42.363.618.59.2110.0 1991 0 1.429.426.115.172.0 1992 0.86.713.346.312.279.3

1994 0.213.20.213.6 1995 13.65.919.4 1996 0.05 1.71.7 1997 1.50.05 1.12.6 1998 2.20.01 15.7* 1.919.7

* incl. 6.2 tons purchased by a factory vessel.

132 Belugas in the North Atlantic and the Russian Arctic Therefore a high option for the catch statistics ciencies, one of which is the fragmentary re- after 1954 (applied to the medium option) in- porting from Qaanaaq. The only correction that cludes a correction of the harvest in northern can be applied to the reported catches is the ad- municipalities (Qaanaaq and Upernavik) of dition of estimated catches for Qaanaaq in the 1.10 of the catches until 1995 and a correction years without reports for that area. The lower factor in all other areas of 1.30 to adjust for and upper 95% confidence interval for the years losses during the catch operations. with catches in Qaanaaq (see Materials and Methods) were therefore added to give a low Ice entrapments and a medium option for corrections, respec- Data on ice entrapments were extracted from tively (see Table 5). For the central and south- Siegstad and Heide-Jørgensen (1994) and an ern areas the low and medium options listed in option for subtracting the entrapment mortality Table 5 are the same as the reported catches. from the reported catches is given for the period 1954-1999. For the period 1862-1953, entrap- All entrapments involving belugas took place in ment mortality is reported inconsistently and is Disko Bay. In January 1955 there was a large therefore not subtracted from the reported ice entrapment during which more than 3,000 catches. belugas were killed (Table 1). Since the report- ed catches came from a settlement close to the RESULTS site of the entrapment and the catches were tak- en in winter it is assumed that all catches in Period 1862-1905 Disko Bay reported from that year were whales Statistics were available for some of the years taken in the ice entrapment. In 1960 an entrap- from Sisimiut and south (1874-1890) where all ment in Disko Bay included 100 narwhals or catches can be considered to be belugas (Table belugas and it is thus assumed that 50 belugas 3). For the areas north of Sisimiut annual catch were taken in the entrapment. In March 1967 data were available for 1862 through 1877 but some belugas were entrapped and 50 reported only for belugas and narwhals combined. Some catches are attributed to ice entrapment mortali- statistics on average catches over periods with- ty. In February 1968 an entrapment was report- out annual reports are presented here as a sup- ed and the catch for that month of 234 belugas plement to the annual reports of catches (see was attributed to ice entrapment. Table 4). Two ice entrapments are known from Disko Bay 1898 and 1899 involving either bel- Period 1975-1985 ugas or narwhals (Table 3). This was a period where an increasing propor- tion of catches were not reported and where Period 1906-1951 corrections for underreporting are needed. The Catches that were reported for the combined pe- annual catch in Qaanaaq has previously been riod 1926-29 in Kangersuatsiaq (in southern estimated to be 50/yr for 1975-1977 and the dif- Upernavik District) were spread over the four ference between the reported catches and this years in question (Table 3). estimate has been added to correct for underre- porting (Kapel 1983). There are no estimates of All entrapment records were from Disko Bay unreported catches for 1978-1982. For 1983- but only the reports from 1906, 1915 and 1951 1985, estimates of underreporting have previ- were specified as belugas, whereas the others ously been added to the reported statistics for may have included both belugas and narwhals. some of the municipalities (Born and Kapel 1986, The catch in the ice entrapment in 1943 might Born 1987, Table 1 column ‘Underreporting’). be reflected in the statistics for 1942 but several of the other ice entrapment catches do not match In January 1976, 653 belugas were reported a similar number in the catch statistics (Table 3) caught in an ice entrapment in Disko Bay. In 1982, 50 to 200 belugas were taken in Disko Period 1954-1974 Bay in an entrapment in February and we arbi- This period is probably the most reliable of the trarily assume that 100 whales were taken from catch series, but still there are some clear defi- this entrapment. In February 1984, 200 belugas

133 NAMMCO Scientific Publications, Volume 4 ments ). The catches es before 1906 are cor- ørgensen 1994). ørgensen N’ and ‘Sisimiut’, that 75% ° N ° ments of 66 N’ including Qaqortoq, Paamiut, Nuuk and Maniitsoq, ‘Sisimiut’ (included in the area ‘South’ 1), ‘Central’ Table in ° N ° of 66 . Catches of belugas in West Greenland between 1862 and 1951 compiled from Winge (1902), Anon. (1944) and Heide-Jørgensen (1994 Anon. (1944) and Heide-Jørgensen (1902), Winge Greenland between 1862 and 1951 compiled from West . Catches of belugas in le 3 ear South Sisimiut Central North Total Ice entrap- Year South Sisimiut Central North Total Ice entrap- are divided into three areas; ‘South of 66 including Disko Bay and ‘North’ including Uummannaq, Kangersuatsiaq and Upernavik. No data are available from other areas. Catch rected for the combined reporting of narwhals by assuming that 100% catches were belugas in areas ‘South 66 Tab Y 186218631864186518661867186818691870 1361871 1301872 2111873 98 106 1071874 215 2151875 234 288 136 237 2801876 166 154 426 4721877 409 180 242 7341878 317 96 369 6521879 83 307 94 248 468 592 169 1913 308 319 308 1914 832 153 249 264 218 198 657 1915 172 240 596 205 625 1916 178 134 670 149 505 1917 116 500 513 141 1919 733 450 413 1920 806 150 1,115 1922 1,100 1923 200 1924 1,380 652 1925 1927 874 592 1926 1928 950 1929 832 950 700 1,500 40 100 100 76 1930 50 596 931 60 670 600 1932 33 25 25-33 450 210 1,140 1,455 100 200 100 636 425 1,050 436 1,436 874 1,050 40 1,336 1,961 1,512 40 536 183 311 575 823 344 1,006 615 were belugas in the ‘Central’ and ‘North’All entrapments were from the ‘Central area’ areas (see text). (see Siegstad and Heide-J

134 Belugas in the North Atlantic and the Russian Arctic ments N ° ments of 66 N ° of 66 . continued le 3 ear South Sisimiut Central North Total Ice entrap- Year South Sisimiut Central North Total Ice entrap- Tab Y 18801881 6941882 4871883 4461884 236 393 1401885 291 1571886 672 1481887 4171888 71 445 1211889 317 2401890 3301898 94 4231899 118 6941906 74 4871907 446 4001908 3931909 90 1091910 291 672 1933 1611911 417 1934 98 1935 300 445 317 1936 150 25 330 1937 1938 9 423 1939 6 13 1940 4 1941 47 8 400 65 6 240 1942 41 5 109 1943 2 8 22 20 6 161 34 80 1 252 49 196 1944 99 98 1946 300 19 78 3 1945 1947 178 48 36 290 186 196 1948 22 326 1 127 5 1950 27 434 2 139 0 380 100 50 56 490 1951 116 253 20 146 162 11 652 100 56 273 9 780 324 659 207 91 238 189 690 355 190 267 41 98 700 340 413 337 296 122 Some 24 122 17 24 17 173

135 NAMMCO Scientific Publications, Volume 4 Table 4. Average catches of belugas in four areas as reported in Anon (1944). The catches are di- vided into three areas; ‘South of 66o N’ including Qaqortoq, Paamiut, Nuuk and Maniitsoq, ‘Sisimiut’ (included in the area ‘South’ in Table 3), ‘Central’ including Disko Bay and ‘North’ in- cluding Uummannaq and Upernavik. Catches are corrected for the combined reporting of nar- whals by assuming that 100% of the catches were belugas in the areas ‘South of 66o N’ and ‘Sisimiut’, and that 75% were belugas in the ‘Central’ and ‘North’ areas (see text).

Years South of 66o N Sisimiut Central North

1887-1888 - - 350 208 1890-1899 322 96 - - 1892-1893 - - 305 219 1900-1909 219 36 - - 1903-1909 - - 209 203 1910-1919 392 37 336 334 1920-1924 162 47 221 264 1925-1929 46 44 173 293 1930-1934 16 39 260 293 1935-1938 6 51 - -

were taken in an entrapment in Disko Bay and In 1992 a total of 46.3 tons of mattak was pur- again in April 1984, 20 belugas were taken in chased in Upernavik. From that catch, 54 jaws an entrapment in Uummannaq. were obtained and, in Disko Bay, an additional 26 jaws were collected (Table 6). Converting Period 1986-1992 the mattak to the number of whales killed (70 The catch reporting system deteriorated during kg mattak/whale, Heide-Jørgensen 1994) and this period while the economic value of the mat- adding the 26 jaws gave a total of 687 belugas, tak increased. Nothing is reported from Qaanaaq but there were definitely other unreported or Sisimiut for any of the years but there are re- catches in West Greenland that year. ports from biologists of catches of 2 and 50 bel- ugas in 1990 and 1991 in Qaanaaq (Table 1). In April 1988, 100 to150 belugas were reported taken from an entrapment; therefore 125 belu- In 1986 a biologist recorded a catch of 450 to gas were arbitrarily assigned to ice entrapment 550 belugas in the two northernmost settle- mortality in 1988. In January 1990 about 500 ments in Upernavik (Bodil Deen Petersen in belugas were taken in an ice entrapment in litt. and later confirmed by the senior author) Disko Bay. and the official statistics were therefore correct- ed to 500 for Upernavik in 1986 (see Table 5). Period 1993-1999 In 1993 the new catch reporting system In 1989 the senior author observed catches of (Piniarneq) started to provide data. The mattak 427 belugas in Upernavik and approximately purchases in the northern municipalities (i.e. 30 in Disko Bay. The reported catches were 116 Qaanaaq, Upernavik and Uummannaq) were and 12 for the two areas thus the observed converted into catches of whales under the as- catches were used. Both observations indicate a sumption of an average yielded amount of mat- significant degree of underreporting. tak per whale (Heide-Jørgensen 1994). The minimum catches thus derived from the mattak Similarly for 1990 and 1991, data on catches purchases were generally smaller than the num- recorded by biologists were used instead of the bers from the ‘Piniarneq’ catch-reporting relatively low reported catches. There was a scheme, indicating that not all mattak was pur- large ice entrapment in Disko Bay in 1990 that chased commercially. The exception was the alone accounted for a kill of approximately 600 catch in Upernavik where the mattak purchases belugas. yielded a much larger estimate of killed belugas

136 Belugas in the North Atlantic and the Russian Arctic ) and to all medium op- q, Upernavik and and a and ‘South’ includes low option low

ENTRAPMENTS k ( 974 a correct for lack of catch reports from Qaanaaq and Sisimiut. medium option and a low ). For 1986-1992 a NORTH CENTRAL SOUTH TOTAL TOTAL WITHOUT ICE Catches of belugas in three areas in West Greenland with three options for corrections of catch numbers. ‘North’West Catches of belugas in three areas includes Qaanaa medium option correct for lack of catch reports from Qaanaaq. For 1975-1985 a correction factor unreported catches is applied to Upernavi able 5. Uummannaq, ‘Central’Aasiaat, Qasigiannguit, Ilulissat and Qeqertarsuaq, includes Disko Bay with the municipalities Kangaatsiaq, Sisimiut, Maniitsoq and Paamiut. Last column show the catches with ice entrapments subtracted from Central area. For 1954-1 areas ( tion T 195419551956 951957 311958 351959 186 351960 122 251961 126 421962 126 217 371963 116 135 531964 1,774 101 133 1401965 105 128 141 1,774 2751966 135 128 373 531967 101 223 149 3911968 105 2,306 275 131 142 1821969 135 373 118 2431970 223 391 112 60 180 179 3581971 222 182 118 23 165 4851972 209 243 150 186 357 219 5081973 271 179 256 243 23 93 2371974 256 262 12 166 336 186 219 3161975 357 243 34 214 313 2331976 243 311 95 30 398 231 93 12 1661977 427 289 36 369 248 242 34 214 285 1,0131978 404 394 1,892 42 154 95 398 231 16 268 18 121 1,013 661 385 216 36 369 1,133 261 44 475 188 278 1,983 42 124 42 165 73 451 1,133 328 517 18 1,317 661 318 403 2,553 47 257 362 480 31 442 30 200 55 291 42 521 581 73 1,473 328 51 23 184 512 859 118 362 243 409 50 84 31 453 30 127 268 581 327 533 55 95 229 51 612 953 426 512 234 34 509 209 50 84 127 170 379 471 40 319 334 39 669 1,134 329 452 773 755 345 418 66 34 170 318 168 109 165 666 451 325 65 412 229 247 1,474 331 442 161 415 538 520 521 488 329 1,277 168 191 221 345 44 409 398 614 579 161 243 170 586 533 120 229 331 191 277 167 612 218 409 699 1,368 1,524 440 996 488 170 234 209 334 705 143 509 1,737 670 199 279 121 248 405 368 329 739 669 1,524 345 1,087 221 600 773 99 325 859 1,043 186 888 144 1,085 259 844 1,911 229 331 1,369 412 329 345 118 1,134 488 913 739 455 1,227 187 564 1,176 579 683 950 398 474 229 331 996 153 912 1,454 488 1,442 1,433 1,155 409 440 885 913 655 1,087 670 474 779 1,844 1,085 739 279 1,144 405 739 859 600 1,369 1,176 998 965 823 574 844 546 913 739 856 950 1,226 1,454 683 608 1,081 913 885 1,155 912 728 739 1,144 760 926 770 898 1,133 970 YEAR Low Medium High Low Medium High Low Medium High Low Medium High Low Medium High

137 NAMMCO Scientific Publications, Volume 4 ENTRAPMENTS NORTH CENTRAL SOUTH TOTAL TOTAL WITHOUT ICE . Continued. le 5 Tab 197919801981 3441982 3571983 5791984 353 5361985 382 3491986 612 4321987 563 393 4201988 363 440 5201989 439 692 379 5791990 457 623 412 1411991 623 403 340 445 4511992 682 486 313 356 4901993 244 504 194 450 4051994 548 686 352 677 586 3721995 356 753 177 484 637 2311996 450 268 231 114 526 4191997 780 603 296 484 177 65 291998 484 393 155 125 114 300 1141999 231 495 163 206 545 30 296 858 77 184 108 684 251 29 230 125 114 538 194 102 100 148 77 206 258 100 129 239 30 684 42 26 251 329 121 194 252 162 100 38 50 148 239 168 48 788 77 924 268 50 1,082 301 157 26 889 194 39 326 60 245 130 239 46 957 60 96 1,056 243 301 65 1,211 881 100 645 312 252 245 78 34 69 1,316 108 86 94 125 1,470 1,064 1,079 311 243 46 826 116 391 312 117 1,275 715 134 1,082 140 647 122 120 924 318 682 788 46 94 148 316 116 152 1,146 174 860 908 120 183 857 406 668 978 312 694 122 148 809 183 94 1,109 1,096 833 183 151 127 156 1,386 645 561 905 183 181 192 1,215 812 122 819 596 463 959 1,157 606 127 238 986 798 678 181 238 1,068 46 712 1,434 931 647 618 552 749 165 682 621 644 780 235 816 542 798 46 609 812 668 694 187 833 576 618 747 900 723 744 780 946 561 60 657 1,014 819 338 542 761 596 812 576 959 958 798 712 744 239 749 704 618 784 644 931 390 749 780 798 967 900 542 816 618 239 576 780 747 744 542 1,014 946 311 576 761 744 958 239 704 749 239 967 311 YEAR Low Medium High Low Medium High Low Medium High Low Medium High Low Medium High

138 Belugas in the North Atlantic and the Russian Arctic Table 6. Number of lower jaws from belugas that have been delivered since the obligatory jaw collection system was put in force in 1992.

1992 1993 1994 1995 1996 1997

Qaanaaq 72 Upernavik 54 339 47 Uummannaq 15 4 9 1 Qeqertarsuaq 24 10 17 10 5 Ilulissat 2 6 67 4 Aasiaat 18 3 10 Kangaatsiaq 19 1 Sisimiut 10 85 37 51 8 Maniitsoq 7 14 41 2 Total 80 477 237 75 113 10 than the reported catches in 1993-1998 (Table 1 belugas) were purchased in Upernavik. This is and 2). not in conflict with the reported catch of 162 whales. A factory ship purchased additionally No reliable data exist on the trade in beluga 6.2 tons of mattak but the locations of the pur- mattak from Uummannaq, Disko Bay, Sisimiut chases are unknown and they can therefore not or south of Sisimiut. Most mattak is sold direct- be compared with the reported catches. For ly by the hunters and no statistics are kept on 1999 the catch statistics only cover the period the amount traded. Also, many fishing boats from January through September. conduct an opportunistic hunt, which makes it difficult to get precise figures on the hunting No ice entrapments of belugas have been re- levels. ported in West Greenland since 1990.

For 1993, the collection of 339 jaws supported DISCUSSION a mattak-derived estimate of 340 belugas in Upernavik (Table 6), twice as many as official- Complete statistics on catches of Monodontids ly reported. from at least the last part of the 19th century probably existed at some time. Some of them In 1994, the mattak purchases in Upernavik in- may still be retrievable from archives whereas dicated a minimum catch of 188 belugas, ex- other parts of the statistics have been lost. A ceeding the reported catch of 98 belugas (Table large number of old archives from southern 1) and again showing a significant underreport- Greenland were lost with the wreckage of the ing. For all other areas, the reported catches ship ‘Hans Hedtoft’ in 1959. The fragmentary were larger than those derived from mattak pur- statistics that are presented here indicate a long chases. When adding the corrected figure for history of relatively high levels of beluga catch Upernavik to the reported catches from all other in West Greenland that probably started before areas an estimated total of 618 belugas were 1862. For population modelling it will be nec- caught in West Greenland that year. essary to interpolate years without reported catches, to spread out the average figures over In 1995 there are records for trade in mattak the years involved and to assume some level of only from the northern area and a total exceed- harvesting before 1862 (cf. Butterworth et al. ing 19 tons was purchased, corresponding to an 2002, Innes and Stewart 2002). estimated hunt of 278 belugas. Despite the incomplete reporting the historic In 1996 and 1997, only small amounts of mat- data clearly shows that some dramatic changes tak were purchased but in 1998 more than 15 in the distribution and occurrence of belugas tons were sold of which at least 9.5 tons (~136 happened in south Greenland in the first dec-

139 NAMMCO Scientific Publications, Volume 4 ades of the 20th century. Catches south of 66° N ical coverage of the catches than previous sys- dwindled during the 1920s probably as a result tems. It is not yet possible to determine whether of over-harvesting. Because of the suspicion this indicates a more complete reporting of that these catches were taken from a now ex- catches, but at least observations from biolo- tinct stock, it seems reasonable to keep these gists in 1994, 1995, 1998 and 1999 in Uper- catches separate from those taken further north navik confirm the information from ‘Piniarneq’ in the present range of belugas. But it is also that catches remained low (<250 whales) in possible that there were other stocks of belugas those years, although underreporting was still a that were depleted by intensive hunting in the problem. last century. For example the settlement of Kangersuatsiaq periodically had high levels of The high option for corrected catch estimates autumn catches (>400/yr during 1926-1932, includes a relatively low (10%) estimate for Heide-Jørgensen 1994), but few belugas have losses in the drive fishery hunt. No studies of been seen or caught there after 1960. losses have been conducted in Greenland but inferences can be made from studies in other ar- For the period after 1954, the localities for the eas. In the western Canadian Arctic, loss rates catches reported in the official catch statistics for belugas were estimated to be about 40%, are registered by the settlement/town where the corresponding to a catch correction factor of hunters live. However this is not necessarily 1.67 (Burns and Seaman 1986). For narwhal where the catches took place, which introduces hunting in open water in Canada, Weaver and a bias, especially in the southern areas where Walker (1988) reported loss rates between 32% hunters make hunting trips to northern whaling and 55%, or catch correction factors of 1.5 to grounds. Thus catch reports from the southern 2.2. Roberge and Dunn (1990) reported catch municipalities may include catches from further correction factors for narwhals in Canada rang- north in other municipalities. This bias has in- ing from 1.11 in open water to 1.41 at ice cracks creased in importance with the widespread use and 1.56 at the floe edge or ice edge. The esti- of motorized vessels since the 1960s. No cor- mate assigned here to losses in the non-drive rections could be applied for this bias but the fishery (catch correction factor of 1.30) in pooling of the catches into larger areas reduces Greenland covers both the open water hunt and its importance. the hunt from ice cracks and the ice edge. It also covers the open-water hunt in late autumn just The use of the average of catches in one decade before freeze-up, which is a period where loss to adjust for underreported catches in a subse- rates have not been estimated. If anything the quent decade could mask both trends and the correction factor of 1.30 applied here is nega- occasional large catches. There are, however, tively biased. no reasons to believe that there have been long periods without catches in Qaanaaq and The two options for correcting for underreport- Sisimiut since both hunting effort and the mar- ing increase the average catches of belugas ket value of the mattak has steadily increased since 1954 by 3.3% for the low option and by over the years and substantial catches are re- 15.5% for the medium option. The high option ported in years with statistics. that includes losses increases the catches by 42.0%. For modelling the dynamics of the pop- In the 1980s and 1990s a large proportion of the ulation an average of the medium and the high catches have been estimated from diverse options has been used (Alvarez and Heide- sources. The worst years are 1991 and 1992 Jørgensen MS 2000, Butterworth et al. 2002). where no official statistics are available, but be- This compromise increases the average catch cause of the arbitrary collection of official data by 28.8%. in all years, an unknown and possibly very large fraction of the catches is not reported (e.g. If the whales killed in ice entrapments are re- no catches in Sisimiut from 1986-1992). The moved then the catch estimates corrected for introduction of a new catch recording system, losses and underreporting (high option) is ‘Piniarneq’ seems to provide a wider geograph- 28.4% larger than the reported catches in the

140 Belugas in the North Atlantic and the Russian Arctic same period. Choosing the same compromise underreporting were exceptional years because between the medium and high options as above they are from periods when the reporting gives an average correction excluding ice en- scheme was undergoing change. Similarly it trapment mortality of 16.7%. can be argued that the loss rates in Greenland are lower than those reported for Canada, be- Since quantitative data on underreporting are cause a larger proportion of the hunt takes place not available for most years and data on losses in the open water season and because hunters in have not been collected in Greenland, qualified Greenland may be more skilled. Thus until guesses of actual catch levels have been re- more accurate information becomes available, quired. The corrections proposed here are clear- the relatively conservative corrections of catch- ly in the lower range of the possible values and es used here seem most appropriate. even the high option may be negatively biased. Data from some years indicate that less than half of the catches were reported, and the ap- ACKNOWLEDGEMENTS plied loss rates are certainly in the lower range of published values for other areas. It can, how- Two reviewers kindly provided comments to ever, be argued that years with observed severe this manuscript.

REFERENCES

Alvarez-Flores, C. and Heide-Jørgensen, M.P. (MS) 2000. Assessment and future harvest options for belugas (Delphinapterus leucas) in West Greenland. Working Paper SC/8/BN/10 for the NAMMCO Scientific Committee.

Anon. 1944. Beretninger vedrørende Grønlands Styrelse. Sammendrag af statistiske oplysninger. [Report concerning Greenland’s governance. Summary of statistical information] Schultz A/S, København 624-632. (In Danish)

Anon. 1953-1971. Sammendrag af Grønlands Fangstlister. [Summary of Greenland’s catch statis- tics] Ministeriet for Grønland, Copenhagen, Denmark. (In Danish)

Born, E.W. 1987. Denmark. Progress report on cetacean research June 1985 to June 1986. Rep. int. Whal. Commn 37:166-168.

Born, E.W. and Kapel, F.O. 1986. Denmark. Progress report on cetacean research June 1984 to June 1985. Rep. int. Whal. Commn 36:151-153.

Burns, J.J. and Seaman, G.A. 1986. Investigations of belukha whales in coastal waters of western and northern Alaska. II Biology and Ecology. Alaska Department of Fish and Game, 1300 College Road, Fairbanks, Alaska 99701.

Butterworth, D.S., Plagányi, É.E. and Geromont, H.F. 2002. Resource assessment and projections for the belugas off West Greenland using the population model of HITTER-FITTER. NAMMCO Sci. Publ. 4: 211-224.

Heide-Jørgensen, M.P. 1994. Distribution, exploitation and population status of white whales (Delphinapterus leucas) and narwhals (Monodon monoceros) in West Greenland. Meddr Grønland, Biosc 39:135-149.

141 NAMMCO Scientific Publications, Volume 4 Heide-Jørgensen, M.P. and Reeves, R.R. 1996. Evidence of a decline in beluga, Delphinapterus leucas, abundance off West Greenland. ICES J. Mar. Sci. 53:61-72.

Heide-Jørgensen, M.P and Acquarone, M. 2002. Size and trends of the bowhead whale, beluga and narwhal stocks wintering off West Greenland. NAMMCO Sci. Publ. 4:191-210.

Innes, S. and Stewart, R.E.A. 2002. Population size and yield of Baffin Bay beluga (Delphinapterus leucas) stocks. NAMMCO Sci. Publ. 4: 225-238.

Kapel, F.O. 1977. Catch of belugas, narwhals, and harbour porpoises in Greenland, 1954-75, by year, month and area. Rep. int. Whal. Commn 27:507-520.

Kapel, F.O. 1983. Catch of belugas, narwhals, and harbour porpoises in Greenland, 1954-75, by year, month and area. Rep. int. Whal. Commn 27:507-520.

Kapel, F.O. 1985. Denmark. Progress report on cetacean research June 1983 to May 1984. Rep. int. Whal. Commn 35:162-165.

Kapel, F.O. and Larsen, F. 1984. Denmark (Greenland) progress report on cetacean research June 1982 to May 1983. Rep. int. Whal. Commn 34:191-193.

Roberge, M.M. and Dunn, J.B. 1990. Assessment of the subsistence harvest and biology of narwhal (Monodon monoceros L.) from Admiralty Inlet, Baffin Island, N.W.T., 1983 and 1986-89. Can. Tech. Rep. Fish. Aquat. Sci. 1747.

Siegstad, H. and Heide-Jørgensen, M.P. 1994. Ice entrapments of narwhals (Monodon monoceros) and belugas (Delphinapterus leucas) in Greenland. Meddr Grønland, Biosci. 39:151-160.

Weaver, P.A. and Walker, R.S.. 1988. The narwhal (Monodon monoceros L.) harvest in Pond Inlet, Northwest Territories: hunt documentation and biological sampling, 1982-1983. Can. Man. Rep. Fish. Aquat. Sci. 1975, 26 pp.

Winge, H. 1902. Grønlands Pattedyr. Meddr Grønland 21:319-521.

142 Belugas in the North Atlantic and the Russian Arctic