NAT. NAT. HIST. BULL. SIAM Soc. 52 (1): 81-109 ， 2004

MEKONG RIVER PANGASIIDAE CATFISH MIGRATIONS AND THE KHONE FALLS WING TRAP FISHERY IN SOUTHERN LAOS

lan lan G. Baird 1 ，2 Mark S. Flahe 吻 2 and Bounpheng Phylavanh 3

ABSTRACT

Laos Laos is heav i1 y dependent on natur a1 resou 町 es to support the livelihoods of the bulk of its its human population. Wild captur 巴 fisheries in the Mekong River and its tributaries are p紅世 cularly important for providing rur a1 Lao people wi 白 income and are their main source

of 戸。館副.百 lis article eX .!Illl ines 飢 important Mekong Ri ver fishery in the ex 回 me south of Laos Laos for migratory pangasiid ca ぽishes and other species at the begiruting of the annu a1 rainy sω80n. Fo 町 y伺 rs of catch-effort fisheries da 飽 for a pair of large wing traps are presented. Al血 ough over 100 0曲er species 釘官 caught in 血.e fishery ，血 .e pangasiid catfish Pangasius conchophilus conchophilus w 錨 byf. 紅白e dominant species in catches. Peak catches for the wing 回， p fishery 釘 e not correlated wi 白 lunar cycles ， Catch data ，and ‘loc a1 ecologic a1 knowledge' of fishers ， suggest 白紙 these fish migrations are associated wi 血 rising river levels at 白 e beginning of the

rainy rainy season. Changes in hydrologic a1 conditions in the Mekong River and its 1紅 ge 凶butaries caused caused by the ∞isstruction of large hydroelec 凶c dams could seriously impact critic a1 1y important fish fish migrations and associated fisheries.

Key words: artisanal ，Cambodia ，capture fisheries ，fr 官shwater fisheries ，s住 'addling stocks ，Laos.

問 TRODUCTION

In In Laos and throughout most of the Lower Mekong River Basin ， fishing and farming in in small subsistence-oriented communities remain the way of life for most people ，and fish and other aquatic products are 也e most important source of animal prote 泊. As Laos is a landlocked nation ，the Mekong River and its tribut 釘 ies 紅 e the ma iil. source of wild cap 伽re fisheries. fisheries. Many local activities and traditions-rice cultivation ， fish h 訂 vesting ，water festivals- are are timed to 出.e seasonal rise and fall of the Mekong. Every ye 釘 around May 由e Mekong River begins to swell as snow melts in the Tibetan mountains and ，more 加 portantly ， monsoon rains 釘 rive 泊 the middle and lower p 紅 ts of the basin. As water levels and cu 汀'ents 加cre 出 e，many fish species become esp 回 ially active ，and most exhibit migratory behavior of v 釘 ying degrees (BAIRD ET AL. ，2003; BAIRD ET AL. ，2001B; BAIRD ，2001; BAIRDETAL. ，1999A; RA 町 BOTH ，1996; S町 GHANOUVONG ET AL. ，1996; ROBERTS & BAIRD ，

lp.O. lp.O. Box 860 ，Pakse ，Lao PDR ，ianbairdialshaw.ca 2Geo graphy De partment ，University of Victoria ，Vic 刷 ia ，B.C. ，C 四tada 3Forestry 3Forestry Section ，Agric 叫.ture and Forestry Di vision ，Ch ampasak pr 制面白， LωPDR Received Received 3 Ocω 加 2003; accepted 10 September 20 倒.

81 81 82 82 IA N G. B A IRD ，M A RK S. FL A H ERTY AND B OUN PHE NG PHYLAV AN H

、 r 、『』，、-『，， ，‘ zzro China ' ，，‘ .... .v .... -( -Av 『、 ， 、、 、 J ) / ・ Vietnam f 、 、、、、ノ

Burma (Myanmar) • Hanoi

/‘ /‘

J 、、 r ‘- Laos Gulfof t Tongken 、〆 、〆

j

~ 〆ノ‘

Thailand Thailand

• Bangkok

Gulfof Thailand Thailand

。 100

Fig山 巴 1. Th es tu c1 y area: Khon g Di strict ，C hamp as 討( Provinc e， Southern Lao PDR MEKONG Rl VER PANGASIDAE CATFISH MIGRAT IONS 83

1995; 1995; ROBERTS & WARREN ， 1994; ROBERTS ， 1993). Rural villagers living near the mainstr 巴am Mekong River's Khon 巴 Falls (Fig. 1 & 2) hav 巴 long be 巴nawru ・巴 ofth 巴 mlgratory nature nature of a lru- ge numb 巴r of fish sp 巴ci 巴s. This includes ， especially ，出巴 catfish family Pangasiidae ， since many of the specie s in this family migrate past th 巴Kh on 巴Falls at the beginning beginning of the rainy season (BAIRD & FLA 旺 RTY ，2004A ; HOGAN ET AL.， 2004; BAIRD ET AL.， 200 lB; B AlR D ， 2001; BAIRD ET AL .， 1999; SlN GHA NOUVONG ET AL. ， 1996; ROBERTS & BALRD ， 1995; ROBERTS ， 1993). Throllghout 出e world ， ru- tisanal fishers have designed fishing geru's to suit their particulru' set set of environmental and socio 巴conomic condition s. Ov 巴lg 巴nerations ， the people in the Khone Fa ll s area have developed an ingenious method for harvesting migratory catfish 一一 the the wing trap ， or '!i' in Lao. Th 巴se impressive looking immovable and V-shaped wooden and and bamboo structures can be up to 14 m Iong and 2 m wide ， not including th 巴 out 巴r wings. The Khone Falls area is th 巴 on1y section of th 巴4，200 km mainstream Mekong River wh 巴re ge010gica l and hydrologica1 conditions are suitab le for building th 巴se espec ia Il y designed fishing fishing devices (BAIRD ， 1998; CLARIDGE ET AL.， 1997; ROsERTS & BAIRD ， 1995; ROBERTS ， 1993) ， although smaller versions are used in tributari 巴s and fI oodplains in th 巴 r巴glO n (CLAR lD GE ET AL .， 1997; DEAP ET AL .， 2003). The Ii fishery is of critical importance to people people in the sO llth ern part of Kh ong District ， and particularly the villages sitllated around the the Kh on 巴Falls. Hundreds of famili 巴s rely on wing traps to provide them with a lru- ge part of of their annllal income ruld food supplies (B AlR D ET AL. ，200 lB; BAIRD ET A L.， 1999A ; BAIRD ， 1998; BAlRD ET AL. ， 1998; SlN GHA NOUVONG ET AL .， 1996; ROsERTS & BAIRD ， 1995; 1995; ROBERTS ， 1993).

Don Don Do n Loppakd l Chome Don Don t:~uay 百 i 7

6 ムl

協徽 S 回 son 拘 inundated nternational _ __ International bound ary Village s an 81 an DonTh ol alh i 10 BaDn onI -$om 2 BaDn onSang 11 8a nD on Sa hong 3 8a nOo Dn Tetok 12 BaHn oua$ada m Ban 4 Ban Do Dn elOke 13 BaDn onPh ap heng CAMBODIA 5 8a Dn onEn 市 4 Ban HangKh one 6 Ba Dn on Tan Tok 15 B町村 ang Sadam 7 Ba Dn oTn anQke 1 6 8an Va un Gn ang 8 Ba nKho neTai 17 Ban 000 Langa 3km 9 8a nKho neN eua 18 Ba nVe uKn h創刊内

Fi gur e 2 Th e Khone Fall s ar ea ， Khong Di strict ，Champa sak Pro vince ，Southern Lao PDR 84 IAN G. BAIRD ，MARK S. FLAHERTY AND BOUNPHENG PHYLAVANH

Large Large rivers are some of the least known ecosystems (BROC 阻 LMAN ，2002) ，and the Mekong River is no exception. In fact ， very little quantitative fisheries data 合om the Mekong River has been published to date. Th is paper presents an analysis of four years (1 994 ，1995 ，1998 and 1999) of fish catch data for two parallel set wing traps situated at the the southem tip of Nok Kasoum Island ，Khong District ，Champasak Province ，southem Lao PD R. This research is p制 of an ongoing effort to document the m 司or artisanal fisheries fisheries conducted around the Kh one Falls in southem Laos ，and associated fish migrations (see ，also ，BAIRD & FLA 田 RTY ，2004A; BAIRD ET A L.， 2003; BAIRD ET AL. ， 2001B).

STUDY AREA

百le Siph 釦 done (or 4，000 islands) Wetland area in the mainstream Mekong River is mainly mainly situated in Khong District ，Champasak Province ，and is a complex and highly diverse diverse ecosystem (See BAIRD ET AL. ， 2003; and DACONTO ，2001 for details regarding the geography geography and people of the study area). Th e Khone Falls represents the southem extent of of Siphandone (Figs. 1 & 2). τbe li fishery is of great importance to villagers living in Ban Hang Khone ， the village where 白is study is based. The southem end of Nok Kasoum Island ，which is less than 2 km northeast of Ban Hang Kh one ，is the site of several important wing traps belonging to villagers villagers from Ban Hang Khone ，Ban Kh one Neua and Ban Don Som (BAIRD ， 1998; ROBERTS & BAIRD ， 1995) (Fig. 2). It is also the site of important fence-filter (tone) traps for for catching migrating small cyprinids during the dry season (BAIRD ET AL. ， 2003). Large Large quantities of fish are caught during peak fishing periods at Nok Kasoum Island (BAIRD ET AL. ， 2003; BAIRD ， 1998; ROBERTS & BAIRD ，1995). 百le strategic location of Nok Kasoum Island ，combined with the relatively large catches from areas adjacent to the island island and the proximity of the area to the project's base at Ban Hang Kh one ，made it an ideal ideal site for studying the wing trap fishery.

THE WING TRAP (L/) FISHERY AT THE KHONE FALLS

Wing traps are commonly used at the edges of rice paddies and in streams throughout Laos and other parts of mainland Southeast Asia (DEAP ET AL. ， 2003; CLARIDGE ET A L.， 1997). 百le li fishery examined in this research is ，however ， fundamentally differen t. It is not not found anywhere else in the mainstream Mekong River outside of the Khone Falls area ， because because this area is the only part of the river that has the physical and hydrological conditions conditions requisite for constructing and operating this particul 訂 type of fishing device (ROBERTS & BAIRD ， 1995). Th e li traps used at the edge of paddy fields and in streams 紅 e smaller 白血 those used at the Kh one Falls ， are shaped differently ，and catch different species species of fish. Th ey generally target fish at the end of the rainy season as they move out of of seasonally inundated areas. The li traps at the Khone Falls mainly catch fish at the very beginning beginning of the monsoon season as they migrate up and down the Mekong River (DEAP ET AL. ， 2003; BAIRD ， 1998; CLARIDGE ET A L.， 1997; ROBERTS & BAIRD ， 1995; ROBERTS ， 1993). 1993). MEKONG RIVER PANGASIDAE CAT 下ISH MIGRA TIONS 85

Each April at the height of the dry season ，villagers construct wing 仕'aps of various sizes sizes and designs using locally obtained wood ，bamboo ，rattan and vines. Each trap is made to to 白血ep 紅白ular physical and hydrological conditions of the site ，refl 民 t凶g 血 e significant amount of local ecological knowledge that villagers hold. Th ese traps are built mainly to target target large schools of pangasiid catfish 血at migrate up the Mekong Ri ver 合omCambodia each each year ， although a wide variety of other species 紅 'e also harvested (BA 限 D ， 1998; ROBERTS ， 1993). Migrators swim up 出.e many channels and rapids that make up 白e Kh one Falls ， but most channels cannot be easily ascended due to the impassible rapids and waterfalls. 百lerefore ，出 ey must move up and down the channels until they find one 出at they can get up. up. On ly a few ， the most notable being Hou Sahong and ， to a lesser extent ，Hou Sadam are are easily passable ye 釘 round (BAIRD & FLAHERTY ，20 倒 A; BAIRD ET AL. ， 2003; BAIRD ET AL. ， 2001B; ROBERTS & BAIRD ， 1995). Villagers Villagers have determined 白紙 the most effective way of catching the migrating fish is is to position 白e traps in the channels so 白紙白e fish can be caught when they are mov 泊g downriver. 百le migrating fish reverse direction when they cannot pass through a channel 白紙 has a high waterfall or steep rapid at its upper end. Some returning fish 紅 e caught in large large quantities over short periods ， while others conduct more pro 位acted migrations (BAIRD ， 1998; 1998; S町 GHANOUVONG ET AL. ， 1996; ROBERTS & BAIRD ， 1995). Lar ge numbers of w 泊.g traps are operated in the Kh one Falls ， but it w 剖 decided 出at it it would be most e百'ective to concentrate the data collection efforts on one key wing trap fishing fishing operation. Accurately monitoring more sites for the full duration of the fishing se 出 on wo u1 d have required much more pr 吋l ect labor than w 槌 av ai1 able. We were fo 此unate h 血at the 血ird author is the son-in-law of one of the co ・owners of a good w 泊g trap site on Nok Kasoum Island. His in-laws are 針。 m Ban Hang Kh one while the other owner lives 泊 Ban Kh one Neua village. 百le two communities are siωated at either end of the 5・・ km long long Kh one Island ，which straddles the Khone Falls at its cen 回 (Fig. 2). 百le wing trap fishery 泊 the Khone Falls area is subjectωlocally developed tenure arrangements 伽.t allow for long-term private ownership of particular trap sites by fishers (BAIRD ET AL. ， 2003; BAIRD ET AL. ， 2001B; ROBERTS & BAIRD ，1995). 百le tenure system for for wing traps in the Nok Kasoum Island 釘 'ea is well established ， and is widely respected by local people. In In recent ye 釘 s，most of the wing traps in the area have been built in essentially the same places and in 血e same ways. For the most p釘 t， the same people have operated 出e 回 ps.. 百lere have been no new en 出es 泊to 出 e li fishery 泊出e Nok Kasoum Island 釘 ea for for at least 15 ye 紅 s. However ， the fishery experienced a period of expansion in the 1970s and 1980s ，when villagers from Ban Hang Khone and other communities in the Kh one Falls 釘 ea established a number of new 住aps. Increased acc 四 s to markets for fish was probably probably the main factor motivating people to establish new trap sites. Th ere are a dozen or or so traps of various shapes ，sizes and efficiency levels in the Nok Kasoum Island area ， and probably more than 100 throughout 血e Khone Falls. Cylindrical current 住aps ，called ‘chip' 泊 Lao (see BAIRD ET AL. ， 2001B; BAIRD ET A L.， 1999B; CLARIDGE ET AL. ，1997) ， are also also used 泊出e Nok Kasoum area during this season ， but are less important 血佃 wing traps. traps. 86 86 lA N G. BAIRD ，MARK S. FLAHERTY 刷 D BOUNPHENG PHYLA V ANH

METHODS

Data Data collection at the wing trap site at Nok Kasoum Island involved monitoring two immovable adjacent traps. Both were 14 m long 叩 d2m wide ， not including the outer wings. 百le traps were built using similar materials and design ， and were placed 泊 the s但 ne location location each year. Al though they 紅 e used together as a single fishing operation ， the trap closest closest to shore has two owners and the outer trap has two different owners. Th e owners of 白e outer trap ，however ， are closely related to one of the owners of the inner trap. Each ye 紅 five or six families ，two of whom 釘 e the owners ，jointly operate the 住aps for the approximately approximately two-month fishing season. 百 le consistent design and operation of the fishery

facilitated facilitated the collection of seasonal data that 釘 'e directly comparable from ye 紅 to year. Neither Neither of the two traps were built in 1996 or 1997 ，however ， because the Mekong River' s waters waters rose faster than expected ，and 出e operators were unable to complete the 回 ps before before the current became too strong for construction to continue (BAIRD ，1998). It typically takes takes about one month of labor by several individuals to build a single large li 住ap. Animist Animist ceremonies are conduct 疋d after each season ， as a show of respect for the spirits who are believed to determine catch levels. Th e vast majority of fish caught in the li fishery are landed during the night. 百tis necessitated necessitated stationing data collectors at 血e 位aps every night over the course of each fishing fishing season. Monitors were also stationed at the traps during the daytime ， since some fish fish are caught then. Small huts built next to 白e 仕aps provide a place for the fishers to sleep sleep during low fishing periods.τbe species and weights of all fish caught during each season season were recorded. Data collection began around mid-May and continued un ti1 fishing ceased 泊 June or July ， depending on hydrological cycles. A season ends for a particular 位ap when it is either washed away by strong cu 町 ents or totally submerged by rising water. Occasionally ， submerged traps 訂 'e not washed away ，加 d re-emerge when water levels drop drop at the end of the rainy season. However ，li trap catches 釘 e only marginal at the end of 血e monsoon season ， even if a 位ap survives the rainy season.

Catch-per-unit 旬 e妊ort ， or CPUE ， is an important tool for assessing and monitoring fisheries ，al 血ough the method has various weaknesses (see BAIRD ET AL. ， 1999B; Cowx ， 1995). 1995). Although we rarely use the term CPUE in 由is paper ，readers should be aware 白紙 since since the two traps monitored are set in one location for the whole season ， they essentially 官 sh' 24 hours a day throughout the fishing season. Th erefore ， the CPUE for the li fishery

at at Nok Kasoum is the total catch for an hour ， day or season ， depending on how one w 佃 ts to to calculate CPUE ， divided by the number of traps monitored ，two ， or the number of fishing ‘fishing operations' monitored ， which was one. Thi s means that the daily catch rates provided provided here are essentially CPUE values for the two 位aps. Fish Fish weights were recorded using the following scales:

1) 1) Fish weighing up to 1 kg were weighed on 1・kg scales with 5g increments.

2) 2) Fish weighing 仕om 1 to 5 kg were weighed on 5・kg scales with 20 g increments.

3) 3) Fish weighing 合om 5 to 15 kg were weighed on 15 ・kg scales with 50 g increments. Because Because the li fisheηcatches 1釘 ge qu 佃 tities of fish over short fishing periods 叩 d villagers villagers are generally anxious to either process or quickly sell p釘 t or all of their catch ， lumping lumping large numbers of fish together for weighing was sometimes necessary. One-kg random samples were taken from large batches of bulk weighed fish. 勘'lE KONG RIVER PANGASIDAE CATFISH MIGRATIONS 87

Hi st 'O rical1y ，m 'O ns 'O'O n seas 'O n fish catches have been preserved f'O r use thr 'O ugh 'O ut 血e ye 低Pr eservati 'O n techniques include fermentati 'O n，pickling ， drying and sm 'O king (FAO ， 1999). 1999). Smal1 cypri 凶ds are general1y salted with br 佃 t'O make fish paste (p a dek). Medium sized sized ca 凶 shes 紅 'e 'O ften sm 'O ked 'O ver an 'O pen frre and 1ater s'0 1d as 'pa yang' ，and 1釘 ge cyprinids cyprinids and ca 由sh are s'0 1d 企esh t'O仕組 ers wh 'O send them 'O n ice t'O market by truck 'O r bus. bus. Fresh and sm 'O ked fish are general1y s'0 1d in the pr 'O vincia1 capital 'O f Pakse ，which is ab 'O ut 130 km fr 'O m 出e Kh'O ne Fal1s. Pa dek is either s'0 1d in c1ay jars 'O utside 'O f Kh'O ng ， is is s'0 1d t 'O vi1l agers 合om c'O mmunities in Kh'O ng ， 'O r is c'O nsumed 1'O cally by the fishers themse1ves. themse1ves. F 'O r the purp 'O ses 'O f 血is research ，migrati 'O n is c'O nsidered t 'O be m 'O vements 'O f fish that result 加組alternati 'O n 'O f species between tw 'O 'O rm 'O re separate habitats ，白紙'O ccurs with a regu1 釘 peri 'O dicity ，and inv 'O lves a1 釘 ge pr 'O p'O rti 'O n 'O f 白 .e p'O pulati 'O n (N OR' 叩 CO 百， 1984). 1984). Migrati 'O ns are 'O ften classified acc 'O rding t 'O their tr 'O phic ，repr 'O ductive and dispersal c'O mp 'O nents ，which differ acc 'O rding t 'O sp 即 ies ， the 'O rigin 'O f particular p'O pulati 'O ns ，and size size classes (S 町 GHANOUVONG Ef AL. ， 1996). Th e fisheries data c 'O llected at Ban Hang Kh'O ne have been entered 泊 t'O a re1ati 'O na1 database database management system (RDMS) using Micr 'O s'O ft ⑧ Exce1 and Access. 官 le variab1es rec 'O rded were weights 'O f individual fish species ， the fishing ge 紅 used t 'O catch the fish ， the the number 'O f ge 紅 s used ， the names 'O f the fishers ， the time peri 'O ds fishing eff 'O rt t'O'O k pl 配 :e ，and 血e number 'O fh 'O urs 血at fishingω 'O k p1 配 e.

RESULTS

Table Table 1 presents the t 'Ota1 catch summaries by species and weight f'O r the 1994 ，1995 ， 1998 叩 d 1999 wing trap 'O r li fishing se 出'O ns. Table 2c 'O mbines the results 'O f all f'O ur seas 'O ns. M 'O re sp 民 ies 'O f fish were caught in the wing trap fishery at N 'O k Kas 'O um Island 也m 泊 any 'O ther fishery m 'O nit 'O red 泊白e Kh'O ne Falls area (BAIRD ， 1998). At least 106 fish fish species and 'O ne 1紅 ge crustacean species (Macrobrachium sp.) were rec 'O rded in catches 'O ver 血ef 'O町 years 出at 由.e fishery was m 'O nit 'O red. Small quantities 'O f 'O ther species may als 'O have been caught ， but were n'O t rec 'O rded individually because 'O f being lumped in the ‘miscellane 'O us fish' categ 'O ry during peak fishing peri 'O ds. 官le data indicate 血at the pangasiid catfish Pangasius conchophilus (p a pho/pa ke in La 'O) is the d'O minant species in the fishery ，m 紘泊g up 40.8% 'O f the c'O mbined catch f'O r the the f'O ur seas 'O ns ， with annual landings r叩 g泊 g fr 'O m between 13.7 and 58%. H 'O wever ，泊 1995 ， catches 'O f the smal1 cyprinid Henicorhynchus lobatus were highest by weigh t. During 白紙 year ， P. conchophilus sch 'O'0 1s apparently migrated up H 'O u Sadam and H 'O u Sah 'O ng channels channels (see F培. 2) ，resulting 泊 uncharacteristically disapp 'O inting catches in the N 'O k Kas 'O um Island 紅 'ea (L 'O cal fishers ，pers. comm. ， 1995). P. conchophilus catches were 白 e highest highest by weight during the 'O ther 白 ree seas 'O ns. Pangasius Pangasius conchophilus are caught 泊 large quantities 'O ver very sh 'O rt time peri 'O ds. Large catches always 'O ccur during the night ，泊dicating that the species h部 an 'O cturnal migrati 'O n pattern. Sch 'O'O ls 'O f fish include individuals bel 'O nging t 'O vari 'O us size classes ， alth 'O ugh members 'O f the 0+ class 釘 e by fi 紅白em 'O st abundan t. Wh ile the migrati 'O ns 'O f larger larger individuals may have a repr 'O ductive functi 'O n ，it is unclear why the sexually 加lID ature individuals individuals migrate (S 町 GHANOUVONG Ef AL. ， 1996). H 'O wever ，f 'O'O d st 'O mach c'O n伽 t蜘 dies 88 IAN G. BAIRD ，MARK S. FLAHERTY AND BoUNP 田 NG PHYLAVANH

Table Table 1 Summary of 1994 ，1995 ， 1998 and 1999 Wing Trap (L z) Fish Catches at Nok Kasoum Island ，Kh ong District ，Champ 出 ak Pr ov 泊ce ，Lao PDR

書 La tin Name T_we ゆtω %catch Min (g) Maxω M 醐 ω StD ev (ω 1994 1994 Pangasi ω， conchophilus 1，314 ，590 46 .1 3 40 2，1αm 341 415 2 I Mis伺:11 m 閃 us fish spp. 543 ，020 19.06 20 20 20 Henicorhynchus 3 I Henicorhynchus lobatus 133 ，5 55 4.69 2 48 18 10 Scaphognathops 4 I Scaphognathops bandanensis 117 ，315 4.1 2 10 200 41 35 Puntioplites 5 I Puntioplites falcifer 104 ，580 3. 67 6 84 37 20 6 I Pangasius krempfi 98 ，650 3. 46 1 ，250 7，7' ∞ 3 ，940 2 ，195 7 I Cy c/ ocheilichthys enoplos 79 ，755 2.80 50 3 ，950 529 804 8 I Pangasi ωbocourti 74 ，045 2.ω 2 ，143 287 564 9 I Pangasi ω， lamaudii 66 ，7∞ 2.34 20 3， 2ω 914 712 10 10 Hemibagrus nemurus 45 ，6卯 Lω 25 %。 129 64 11 11 Other (73 脚 i凶+) 271 ，5 53 9.53 Totals Totals 2， 849 ，453 1ω.00 1995 1995 Henicorhynchus 1 I Henicorhynchus lobatus 326 ， 3ω 24.01 8 2活 15 5 2 I Pangl ωius conchophilus 185 ，818 13.67 50 3，4∞ 420 “7 3 I Cosmochil ω， ham ωndi 119 ，690 8.81 25 61 制E日 1，3 23 1 ，462 4 I Pangasius krempfi 103 ，450 7.61 1，5∞ 5，5∞ 3 ，277 1，198 5 I Pangasi ω， lamaudii 94 ，920 6.98 80 4，2∞ 832 640 Miscellaneous 6 I Miscellaneous fish spp. 68 ，000 5.ω 7 I Pang ω， 'ius 初courti 63 ，110 4.64 5 4，9∞ 833 984 8 I Hemibagrus wyckioides 54 ，820 4.03 2∞ 7，9∞ 3. ∞5 2 ，139 9 I Hemibagrus nemurus 49 ，550 3. 65 20 1，0∞ 235 183 10 10 Pangasianodon hypophthalmus 46 ，4∞ 3.41 150 2，5∞ 1，075 ω9 11 11 Oth er (61 speci 凶+) 246 ，962 18 .1 7 Totals Totals 1，3 59 ，080 1ω.ω 1998 1998 Pangasius 1 I Pangasius conchophilus 1，211 ，185 57.94 14 4，5∞ 369 ω4 Henicorhynchus 2 I Henicorhynchus lobat 凶 304 ，155 14.55 6 40 18 10 3 I Cosmochili ω加問ndi 80 ，330 3. 84 10 4 ，600 1，236 1，211 Scaphognathops 4 I Scaphognathops bandanensis 68 ，695 3. 29 10 92 42 21 5 I Cy c/ ocheilichthys enoplas 44 ，010 2.1 1 180 3;αm 1，142 887 6 I Pu nJi oplites falcifer 29 ，110 1. 39 13 120 49 27 7 I Henicorhynch ωsiamensis 28 ，007 1.3 4 9 54 27 11 8 I Pang ω， ius kren 柑 27 ，6∞ 1. 32 2，4∞ 5，αm 3，421 965 9 I H emibagrus nemu 加 25 ，380 1. 21 88 450 195 105 10 10 La biobarbus leptoche i/， 附 23 ，693 1. 13 10 44 22 8 11 11 即断。4 sp 倒 es +) 248 ，195 11. 87 Totals Totals 2，酬，3 60 1ω.ω 1999 1999 Pangasius 1 I Pangasius conchophilus 653 ，3∞ 33 .3 3 188 2，2∞ ω2 710 Scaphongnathops 2 I Scaphongnathops bandanensis 379 ，270 19 .3 5 3 l∞ 47 29 Henicorhynchus 3 I Henicorhynchus lobatus 271 ，645 13. 86 3 40 17 10 4 I Pangasius krempfi 160 ， 5ω 8.1 9 1.αm 8，“ 7 2 ，948 2 ，305 5 I Pangasius bocourti 158 ，印加 8.06 2 ，167 2 ，167 2，167 Cosmochilus 6 I Cosmochilus hom 切ndi 65 ，990 3.3 7 50 3 ，333 1，135 892 7 I Pangasii ωla 糊凶r;; 54 ，650 2.7 9 1ω 3，0∞ 1，164 716 8 I Pangasius mocrone 脚 40" 制 2.06 17 50 37 16 9 I Cy c/ ocheilichthys enoplos 40 ，280 2.06 280 4，1α)() 1，883 913 Puntioplites 10 Puntioplites falcifer 27 ，215 1. 39 7 l∞ 67 28 11 11 Others (49 s陣取+) 108 ，729 5.55 To 旬Is 1，959 ，979 1ω.00 Table Table 2 Combined Summary of 1994 ，1995 ， 1998 and 1999 Wing Trap (Li) Fish Catches at Nok Kasoum Island ，Kh ong District ， Champasak Pr ovince ，Lao PDR

# La tin Name La o Name T_ weight (k g) %ω ，tc: b Min ω M 眠 ω M 醐 E 匂} StD ev 包)

Pangasi 凶， conchophilus papholpa ke 3 ，365 40.76 14 4，5∞ 397 ω2 2 Henicorhynchus lobatus pa soi houa lem 1， 036 12.55 2 48 18 10 3 Mi scellaneo 凶 fish spp. pa louam 611 7.40 4 Scaphognathops bandanensis paplan 578 7. ∞ 3 1，α10 48 88 玄刷回円 5 Pangasius kre 叩， F pa so 邸宅 y hang leuang 390 4.72 1，瓜 10 8 ，667 3， 359 1 ，685 OZO 6 Pangasius bocourti pa houa mouam 3ll 3.77 5， 400 496 929 7 Cosmochilus harmondi pa mak ban 295 3.57 10 6;α10 1， 062 1， 204 8 Pangasi ω larnaudii papeung 221 2.68 20 4，2∞ 927 665 何回〈開何回 Cyclocheilichthys 9 Cyclocheilichthys enoplos pa choklpa choke 207 2.51 50 4，∞ o 923 9f5 10 10 Puntioplites falc 俳r pa sa 初， ng 170 2.06 6 143 55 29 〉M 11 11 Hemibagr ω nemurus pa kot le 嗣， ng 125 1. 51 20 1 ，α)() 187 138 ZO 12 12 Hypsibarbus malcolmi pa pak komlpa pak no 回 t 77 0.93 10 2，6ω 〉臼 13 13 Hemibagrus 町 'ckioides pa kheung 76 0.92 15 7 9 2 436 1 807 ，∞ ， ， HU 刷， pa gnone tharruu お 16ω 14 14 Pangasius macrone 64 0.78 ， 53 162 〉加の〉叶国臼回忌問。 15 15 Bagarius yarrellilspp. pa khe 57 0.69 l∞ 9，5∞ 1， 270 1 ，941 16 16 Henicorhynchus siamensis pa soi ho 雌 po 55 0.67 9 l∞ 28 14 17 17 Pangas 刷出品目 n hypophthalmus pa souay kheo 48 0.58 150 2，5∞ 1，ωo ω2 18 18 La biobarbus leptocheilus pa lang khon 44 0.53 4 5∞ 29 56 19 19 Hemisilurus mekongensis pa 即時deng 43 0.52 63 1;α10 429 216 20 20 Henicorhynchus lobatus and siamensis pa SOI 42 0.51 21 21 Morulius chrysophekadionlspp. paphia 42 0.51 100 4 8 985 m ，∞ 938 〉吋同 22 22 Pangasius pleuro 加enia pa gnone thong 幼 om 40 0.48 5 332 74 46 23 23 Belodontichthys dinema pa khop 35 0.42 75 1，8∞ 747 452 OZ 24 24 Helicophagus waandersi pa nou 32 0.39 50 1 ，833 241 228 臼 25 25 Probarbus jullieni pae 醐 ta deng 20 0.24 5 6，5∞ 174 828 26 26 Cirrhinus microlepis pa phone 19 0.23 35 3，6∞ 687 984 27 27 Wallago attu pakhaa 18 0.22 l∞ 9 ，400 3 ，058 3， 357 28 28 Lo bocheilos melanotaenia pa khiang 18 0.22 5 2∞ 36 27 29 29 Cyprinus carpio pa nal 17 0.21 5 1，5∞ 258 307 30 30 Chitala blanci pa tong koi 17 0.21 20 2，5∞ 745 737 31 31 Paralaubuca typus pa tep 14 0.1 7 3 850 20 90 32 32 Leptobarbus hoeveni pa phong 13 0.1 6 850 1，6∞ 1 ，225 530 、。 E晶 Table Table 2 (continued). 8

# Latin Name Lao Name T_ weight (kg) % catch Min (g) M 鉱 (g) Mean (g) StDev (g) 33 33 Tenualosa thibaudeaui pa makphang 12 0.1 5 3 120 17 20 34 34 Micronema apogon/M. micronema pa nang khao/pa sa ngoua 10 0.1 2 8 2 ，000 533 567 35 35 Barbodes altus pa 凹"a n fai 9.8 0.1 2 1J 300 63 43 36 36 Cirrhinus mrigala pa nang chan 9.0 0.1 1 150 3 ，200 1 ，129 1 ，182 FZ0 37 37 La beo eηthropterus pa va so ω ng 8.9 0.1 1 60 1，∞ o 371 284 38 38 Wallago leeri pa khoun 8.0 0.1 0 8 ，0∞ 8，∞ o 8 ，000 ・回〉見 39 39 Hypsibarbus wetmorei pa pak thong leuang 8.0 0.10 80 1 ，200 655 447 40 40 N otopterus notopterus pa tong na 7.6 0.09 60 300 144 61 U- 41 41 Osphronemus exodon pa men 6.8 0.08 20 1 ，200 456 324 玄〉国間的・明 42 42 M icronema bleekeri pa nang ngeun 5.4 0.07 10 1，2∞ 461 297 43 43 Osteoch i/ us melanopleurus pa khang lai 5.3 0.06 10 2 ，300 240 562 44 44 Kryptopterus spp. pa pik kai 4.9 0.06 10 150 41 25 EE 45 45 Chitala ornata pa tong 幼0 凶 y 4.7 0.06 2 ，650 2 ，650 2 ，650 69 69 61 国 46 46 Amblyrhynchichthys trunca 加 pa ta po 4.2 0.05 5 310 qJ 47 47 Cros$ocheilus reticulatus pa toke thoi 3.9 0.05 5 65 13 8 へ〉 48 48 Bangana behri pa va na no 3.5 0.04 50 725 160 202 20

49 49 Gyrinocheilus pennocki pa ko 3.2 0.04 10 260 125 79 回 O 50 50 Probarbus labe. 制 ajor pa eun 幼 ao 2.8 0.03 60 640 271 246 巴宅回目 51 51 Sikukia gudgeri pa khao na 2.7 0.03 4 100 32 26 52 52 Botia modesta pa mou man 2.7 0.03 8 50 17 B wZO

ma/ 即 10 100 28 17 53 53 La ides hexane spp. pa g ne thong 2.5 0.03 司 54 54 M ekongina erythrospila pa sa-I 2.4 0.03 10 85 43 20 Z4EJ 55 55 Hypsibarbus lagleri pa pakpe 2.3 0.03 25 400 223 141 129 129 154 『〉 56 56 Thynnichthys thynnoides pa koum 2.1 0.03 20 400 ZZ 57 57 Bagrichthys macracanthus pa mak 幼 an mak kheu 2.0 0.02 50 320 145 88 58 58 Hemibagrus 町 cki pa kot mo 1. 9 0.02 950 950 950 。

59 59 Scaphognathops st り， :negeri pa pwn 1. 9 0.02 50 200 107 42 60 60 Mystus singaringa n/ spp. pa kha gneng 1. 6 0.02 20 100 53 26 61 61 Channa striata pa kho 1. 5 0.02 50 900 363 373 62 62 Ompok bimaculatus pa seuam 1. 2 0.01 20 300 112 87 63 63 Cyclocheilichthys repasson pa doke ngieu 1. 2 0.01 20 150 48 34 64 64 Hampala macrolepidota pa sout 1. 0 0.01 5 460 39 98 Table Table 2 (con 出lU ed).

# La tin Name La o Name T_weight (同 % catch Min (g) Max(g) M 倒 I( 紛 StDe vω

65 65 Pang ω， iusp o. 加ranodon pa gnone hang hian 0.80 0.01 250 3∞ 267 29 66 66 Cirrhinus molitorella pa keng 0.80 0.01 20 5∞ 114 173 67 67 Xenentodon cancila pa kathong 0.78 0.01 7 40 21 10 177 177 68 68 Macrochirichthys mocrochirus pa hang pha 0.75 0.01 250 500 375 富田 69 69 Channa moruliuslspp. pa kouan 0.68 0.01 2ω 420 340 113 82 70 70 Pseudomystus siamensis pa khi hia 0.64 0.01 8 67 34 18 71 71 Pristolepis fasciata pa ka 0.59 0.01 5 80 21 21 。自〈恩 72 72 Osteochilus microcepha l. 凶 waandersii pa 幼ang lai gnai 0.57 0.01 20 45 28 12 73 73 Osteochil ω， hasselti pa mak bo ωp 0.44 0.01 10 50 26 14 74 74 Hampala dispar pa sout 0.36 O. ∞ 10 220 63 89 NPP20 75 75 Botia helodes pa kheo μi 0.3 5 O. ∞ 10 45 24 7 76 76 G か'fJ tothorax spp. pa kon 0.34 O. ∞ 5 15 7 3 77 77 Garra fasciacaudo pa khiko 0.3 3 O. ∞ 3 20 8 4 〉臼田〉加の〉同団臼剛山富白河〉国 78 78 Mystacoleucus margina 加 pa lang 聞 m 0.3 3 O. ∞ 7 25 12 5 79 79 Coi 凶， undecimradiatus pa se ω 0.32 O. ∞ 320 320 320 80 80 Macrobrachium sp. (large shrimp) koung gnai 0.3 0 O. ∞ 3∞ 3∞ 300 81 81 Luciosoma bleekeri pa mak vai 0.28 O. ∞ 25 l∞ 47 30 82 82 Raian 昭s gutta 飢S pa sanak noi 0.28 O. ∞ 10 70 31 21 83 83 Mastacemblus ar 刷， tuslspp. pa lat 0.25 O. ∞ 250 250 250 84 84 Cyclocheilichthys armatus pa doke ngieu 0.25 O. ∞ 10 50 27 18 85 85 Cirrhinus jullieni pa doke ngieu pa 0.24 O. ∞ 35 65 50 12 86 86 Set 伊inna mela 即 chir pa meo 0.21 O.ω 10 2∞ 105 134 41 41 87 87 Mystus multiradi 凶， imysticetuslspp. pa 幼a gneng 幼ang lai 0.21 O. ∞ 15 80 24 OZ pa pa fa hanglpa fa lai 0.20 2ω 88 88 Amphotisti ω laosensis O. ∞ 2∞ 2∞ 臼 89 89 Rasbora 叩p. pa Sleu ao 0.1 9 O. ∞ 5 20 9 4 卯 Poropunti ，凶 deauratus pa chat hang leuang 0.1 7 O. ∞ 7 80 25 28 91 91 Sys ω mus orphoides papok 0.1 6 O. ∞ 10 30 22 8 92 92 Osteochil 附， waandersi pald 凶ng lai 0.1 0 0.00 20 80 50 42 93 93 Clarias batrachuslspp. pa douk 0.1 0 O. ∞ lω 1∞ 1∞ 94 94 Onychostoma cf. elongat 即 Z pa 幼iang fai 0.1 0 O. ∞ 10 20 14 5 95 95 Toxotes microlepis pa mong O.ω O.ω 20 25 23 3 0.09 0.09 4 10 7 3 96 96 Opsarius pulchellus pa sleu ao O. ∞ 1.0 h 司 U コ Table Table 2 (con 白lU ed). ト且

# Latin N阻垂 La o Name T_weight (匂} %ω 帥 Minω Max (g) M 則包) StDev (g)

97 97 Kryp ω 'P terus cryp ω 'P terus pap 此 kai 0.05 o. ∞ 50 50 50 98 98 Tetra 似品。， 'n leiuruslspp. papao 0.05 o. ∞ 50 50 50 99 99 Mystacoleucus atridorsalis pa lang nam 0.03 o. ∞ 5 10 7 3 100 100 Crossocheilus siamensis pa kJ 切ng lai noi 0.03 o. ∞ 5 20 13 11 vz 101 101 4 Acantopsis Acantopsis spp. pa hak kauay 0.02 o. ∞ 5 10 s 。・回〉田口・玄〉回開臼・国与国出回 102 102 Tor tambroides pa koua 0.02 o. ∞ 20 20 20 103 103 Botia caudipunctat a/ spp. pa mou mon 0.02 o.ω 初 20 20 104 104 Trichogaster spp. pa kadeut 0.01 o. ∞ 10 10 10 105 105 Parambassis wo 伊:/spp. pa khap khong o. ∞5 o. ∞ 5 5 5 To 凶s 8 ，257 Iω. ∞

44EU

02回

噌呂田

ZOE4F

ミ〉

Z 国 MEKONG RIVER PANGASIDAE CATFISH MIGRATIONS 93 conducted conducted in 出e Kh one Falls area indicate that this species changes its diet considerably by season (BAIRD & PHYLA V ANH ， 1999). Th erefore ， migrations may have a trophic componen t. S町 GHANOUVONG ET AL. (1 996) suggested 出at many of the larger migrating P.conchophilus P.conchophilus probably only exhibit gonadal development after reaching their intended spawning spawning ground. BAIRD & PHYLAVANH (1999) have found individual P. conchophilus in various various stages of reproductive condition 泊 March ， June and August ，indica 由19 血atspaw 凶ng pattems pattems may be complicated ， and spawning may occur more 仕1an once a year. SINGHANOUVONG ET AL. (1 996) found 血at Pangasius conchophilus constitutes a large proportion proportion of the li catch from Hou Som Gnai channel ，a steep gradient channel on 出e east side side of the Kh one Falls. We believe that the species is the most abundant caught by li 回 ps throughout throughout the Kh one Falls. LIENG ET AL. (1 995) recorded P. conchophilus as the 12 由 most common fish species 卸 the bag-net (da i) fishery in the Tonle Sap Ri ver 加 central Cambodia ， albeit albeit only making up 0.18% of the catch. However ，it is unclear how far 白is species migrates migrates or if there is any relation between the populations targeted by the dai fishery in Cambodia and those targeted by the li fishery in Laos. Th e cyprinid minnow Henicorhynchus lobatus (p a soi houa lem) was the second most abundant abundant species in catches by weight ，constituting 叩 average of 12.5% (range 4.7-24%) of of catches. In fact ， the actual total catch for 血is species may be slightly underestimated ， since since some of the fish in 血e ‘miscellaneous fish species' and 'Henicorhynchus lobatus and siamensis' categories were H. lobatus ， along with other small cyprinids ， including H. H. siamensis (p a soi houa po) ， Labiobarbus leptocheilus (p a lang khon) ，Crossocheilus reticulatus reticulatus (p a toke thoi) ，Paralaubuca η 'pus (p a tep) 叩 d Lobocheilus melanotaenia (p a khiang khang lai). Most H. lobatus were caught during the day. 百lere were apparently fewer fewer H. lobatus caught in li 甘aps in Hou Som Gnai compared to the traps located near Nok Kasoum Island (BAIRD ， 1998; SING 臥 NOUVONG ET AL. ， 1996).

Unlike Unlike p佃 gasiid catfishes ，which are migrating upriver when they are caught in wing traps ， Henicorhynchus lobatus are caught as they move downriver to spawn in the Great Lake and other seasonally inundated wetlands in Cambodia. SINGHANOUVONG ET AL. (1 996) and and ROBERTS & W ARR 凹(1 994) reported that villagers 仕om Kh ong Dis 凶ct believe that 出is species spawns above the Khone Falls. 官 le many villagers with whom we have spoken ， concur with these authors 白紙 H. lobatus make grunting sounds when 血ey 釘 e still above above the Kh one Falls. However ， they assert that 由e v. ast majority of H. lobat 凶 spawn downriver downriver in Cambodia. Th is claim is backed up by our field observations that no spent H. H. lobatus 訂 e caught in fish attractant baskets that t釘 'get H. lobatus and other small cypri 凶ds at Ban Hang Kh one below the Kh one Falls in June and July (BAIRD ， 1998). Th e third most abundant species was the medium-sized cyprinid Scaphognathops bandanensis bandanensis (p a pian) ，which made up an average of 7% (range 0.9-19 .4%) of catches. Villagers Villagers believe 由at these fish 紅 'e moving upriver when they 制定 caugh t. L 訂 ge runs 釘 e usually usually landed at the very beginning of the wing trap fish 泊g se 部 on ， before peak P. conchophilus conchophilus runs begin. For example; over 100 kg of the species were landed on May 26 ， 1994. 1994. Migrations app 訂 ently take place during 白e dayt 加 e， as this is when the fish 釘 e caught caught in the li 回 ps (BAIRD ， 1998). Large runs were caught in 1994 ， 1998 and 1999 ， but only only a few fish were recorded in 1995. Th e migrating schools of this species may not have have passed Nok Kasoum Island in 1995 ，or 血ey may have migrated upriver before the li li traps at Nok Kasoum began operating. SINGHANOUVONG ET AL. (1 996) did not record this species species in li catches at Hou Som Gnai ， and it is not found in the lower stretches of the 94 IAN G. BAIRD. MARK S. FLAHERTY AND BOUNPHENG PHYLAVANH

Mekong Ri ver in Central Cambodia. It is believed to conduct annual migrations between the the Sekong ，Se San and Sre Pok b踊 ins ，and in the Mekong Ri ver ， in northeast Cambodia and southem Laos (BAIRD & FLAHERTY ，2004A; BAIRD EI' ι ， 1999A; ROBERTS & BAIRD ， 1995). 1995). However ， migratory pattems are apparen t1 y complex (BAIRD & Fl..AHER TY ，2004A) and are not fully understood (BAIRD &孔 AHERTY ，2004A; BAIRD EI' AL. ， 2003; BAIRD EI' AL. ，2001B; BAIRD ET AL. ， 1999A; WARREN EI' AL. ， 1998). 百 e pangasiid ca 凶 sh Pangasius krempfi (p a souay hang leuang) is believed to be an anadromous species 血at migrates up 企'om the Mekong Delta 泊 Viet Nam to 血.e Kh one Falls 佃 d beyond (BAIRD ， 1998; SINGHANOUVONG EI' AL. ， 1996; ROBERTS & BAIRD ， 1995). Recent Recent examination of P. krempfi olitoths (‘即 stones') 加 d stable isotopes have confmned 血at fish caught at 出e Kh one Falls migrate 企om the South China Sea and the Mekong Delta Delta es 加町泊Viet Nam (H OGAN EI' AL. ， 2004). It was 血e fourth most abundant spec 同 by weight 凶 catches ，m 紘泊.g up an average of 4.7% (range 1. 3-8.2%) of to 凶 landings. It It appe 紅 S 白紙 0凶y individuals in spawning condition migrate up to 血e Kh one Falls (HOGAN EI' AL. ， 2004; BAIRD EI' AL. ， 1999A; BAIRD ， 1998; ROB 臨時&BAIRD ，1995). 百le l釘 gest individual landed 泊血 e li traps w 錨 8.7 kg ， the smallest was 1 kg ，and the me 組 weight weight was 3 .4 kg. BAIRD & PH 礼 AV 釧 H (1 999) exam 泊ed 79 spec 加 .ens of adult P. krempfi krempfi caught ne 釘 Ban Hang Kh one at the beginning of 由e rainy season and found that 56% were females with immature eggs. 官 le rest were mal 回. 94% of 血e spec 耐lens examined had empty stomaches and large amounts of body fat ，a condition common for anadromous fishes ，which typically do not feed 泊 freshwater environments and derive energy energy for migration and gonadal maturation 金om fat deposits dis 住ibuted 白roughout 出e tissues tissues and viscera (BE 悶 ATCHEZ & DODSON ， 1987). Al though 出is species is important to w 泊g 回 p and 1釘 ge-meshed gillnet fisheries ne 紅 Ban Hang Kh one (BAIRD EI' AL. ， 1999A; BAIRD ， 1998; ROBERTS & BAIRD ，1995) ，S 町 GHANOUVONG EI' AL. (1 996) observed on1 y two specimens at Hou Som Gnai between 1993 and 1996. Pangasius Pangasius bocourti (p a houa mouam) is ano 由er p佃 gasiid ca 出 sh believed to migrate upriver upriver during 血is season (BAIRD ， 1998; RA 町 BOTH ，1996 ，S凹 GHANOUVONG EI' AL. ， 1996). It It was 血e fi 佃1 most abundant species by weight in catches ，making up 3.8% (range 0.8-8.1 0.8-8.1 %) of total catches. Although most of the fish caught are medium-to-large size ， very very small pangasiid juveniles tentatively identified as P. bocourti were also caught 加出e wing 甘ap fishery at Nok Kasoum. However ， fmgerlings of Pangasius hypophthalmus ， Pangasius Pangasius larnaudii ， Pangasius conchophilus and Pangasius djambal 訂 'e sometimes mistaken mistaken for P. bocourti 泊 Viet Nam (LENORMAND ，1996). Nevertheless ，RA 到 BO'四(1 996) reported 白紙 P. bocourti spawns at 出.e outset of the flood season ，and 血at juveniles reach about about 5 cm by 凶 d-June. No other juvenile pang 錨 iids were found in catches. 百le cyprinid carp Cosmochilus harmandi (p a mak ban) was the sixth most abundant species species by weight ，m 北 ing up 3.6% (range 1.0-8 .8%) of catches. ROBERTS (1993) suggested 白紙偽is species migrates downriver at the beginning of the rainy season ， but the first author author has observed C. harmandi unsuccessfully a批 mp 出19 to jump up 血e Somphamit Falls ，which is on the west side of Kh one Island (Fig. 2) ， at the beginning of 血e rainy season. 官邸 indicates that at least some ぽ e 甘y泊gωmove upriver at 出at time of ye 釘， but but we do not believe that C. harmandi is a long distance migrator (BAIRD EI' AL. ， 1999A). 百le pangasiid catfish Pangasius larnaudii (p a peung) was the seventh most abundant species 加 catches ，making up 2.7% (range 0.23-7%) of the totallandings. 官邸noctumal migrator migrator is believed to be moving upriver when caugh t. Catches 紅 'e much more pro 住acted MEKONG RIVER PANGASIDAE CATFISH MIGRATIONS 95 compared to those of Pangasius conchophilus. (See ，also ，SINGHANOUVONG Ef AL. ， 1996). BARDACH (1 959) wrote that some sp 配 ies of Pangasius ，including P. larnaudii and Pang ω， ius hypophthalmus ， migrate up the Mekong River during the low water dry season before spawning 泊 the Stung Treng and Kh one Falls 釘 ea in June and July. However ，little is known about the spawning grounds of this species near 血e Kh one Falls ，and fingerlings have not been observed there. 百le cyprinid c釘 p Cyclocheilichthys enoplos (p a cho k/ pa choke) was the eighth most abundant 叩ecies 加 catches at 2.5% ，and the cyprinid Puntioplites falcifer (p a sa 初 ng) was the the n泊出 most abundant at 2.1 %. ROBERTS (1 993) suggested 白紙 C. enoplos migrates downriver downriver at the Kh one Falls at the beginning of the rainy season. However ，we now believe 白紙 both species 紅 'e moving upriver when they are caught in wing traps at 也e beginning beginning of the rainy season. The first author has observed both species 位ying unsuccessfully unsuccessfully to jump up the Sompamit Falls 泊 May ，indicating 白紙 they 釘官官ying to move upriver at that time of ye 低 SINGHANOUVONG ET AL. (1 996) did not record either species species at Hou Som Gna i. Th e bagrid catfish Hemibagrus nemurus (p a kot leuang) ，出e tenth most abundant species species in catches ，made up 1. 5% of landings. Wh ile SINGHANOUVONG Ef AL. (1 996) proposed 白紙偽is species is probably caught when migrating upriver ，we believe that 白紙 at at least some move downriver at the beginning of 出e rainy season.τ 'h is hypothesis is based based on 白e occurrence of an impo 託ant' 初 sone' funnel trap fishery just above the Kh one Falls ，which specifically targets schools of H. nemurus when they move down the Kh one Falls Falls at the beginning of the rainy season (BAIRD ET AL. ， 1999A). Th e cyprinid c釘 p Hypsibarbus malcolmi (p a pak komlpa pak nouat) was the eleventh most abundant species ，constituting 0.9% of the catch. Whi1 e ROBERTS (1993) suggested that that the species migrates downriver at 出e beginning of the rainy season ， the first au 血or has has observed it unsuccessfully attempting to jump up the Somphamit Falls at the beginning of of the rainy season ，indicating 白紙白e species migrates upriver at the beginning of the ramy season. Th e large bagrid ca 出 sh Hemibagrus wyckioides (p a kheung) ， the twel 帥 most abundant species species (0.9% of the catch) is not believed to be highly migratory ， but probably disperses short short distances at the beginning of 白e rainy season (BAIRD Ef AL. ， 1999; SINGHANOUVONG Ef AL. ， 1996). 百 le p佃 gasid Pangasius macronema (p a gnone thamada) (th 註teen 出 most abundant species species with 0.8% of catches) migrates mainly in April before the li 甘'aps at Nok Kasoum begin begin operating (BAIRD ET AL. ，2001B; ROBERTS & BAIRD ， 1995). BAIRD Ef AL. (2001B) documented the fisheries in 血e Hou Sahong channel of the Khone Falls 白紙 specifically t紅 'get migrating schools of 出is species. Migrations Migrations of 1釘 ge sisorid ca 凶 sh Bagarius yarrelli/spp. (p a khe) 釘 'e not well known ， but but the species may conduct short distance migrations wi 白 dispersal and trophic components (BAIRD Ef AL. ， 1999A). SINGHANOUVONG ET AL. (1996) hypothesized 伽 t 白e species migrates migrates up the Kh one Falls through Hou Som Gnai ， although they acknowledged that local fishers fishers local do not consider the species to migrate long distances. Pangasius Pangasius hypophthalmus (p a souay kheo) is another pangasiid catfish believed to move up the Mekong River from Cambodia to the Kh one Falls at the beginning of the rainy rainy se 部 on. Whi le it made up only 0.6% of the total catch ， the fish constituted 3.4% of of landings 1995. None were caught in 1994 and 1999 ，and only one in 1998. Ze b Hogan 96 IAN G. BAIRD ，MARK S. FLAHERTY AND BOUNPHENG PHYLAVANH

tagged tagged a number of P. hypophthalmus caught in the bag-net fishery ne 紅 Phnom Penh in

November 2001 ，叩d was later able to recover the tag of 17 ・kg specimen two months later ， in in Stung Treng ，Cambodia ，300 km up the Mekong River and not far south from 出e Kh one Falls. Falls. Since then tags attached to a number of other P. hypophthalmus caught in the bag- net net fishery have also turned up near Stung Treng (HOGAN ET AL. ， 2004). Th e migratory staωs of three more pangasid catfish caught in smaller numbers in the li li fishery remains unclear. Pangasius pleurotaenia (p a gnone thong khom) may conduct short short distance migrations ， such as one in 1994 documented by S別 GHANOUVONG ET AL. (1996) (1996) from Hou Som Gnai ， but we believe that they do not migrate long distances like some other pangasiid ca 出品目 (BAIRD ET AL. ， 1999A). Helicophagus waandersi (p a nou) may also participate in short migrations ， but we think it unlikely that they migrate long distances distances (BAIRD ET A L.， 1999A). Pangasius polyuranodon (p a gnone hang hian) is not believed believed to move long distances (BAIRD ET AL. ，1999A) ，and since the species is only rarely found above the Khone Falls (ROBERTS ，1993) ，it is highly unlikely that large numbers move up the Falls. Labeo eη thropterus (p a va souang) made up only 0.1 % of catches ，and only small sized sized individuals of this large cyprinid were landed. SINGHANOUVONG ET AL. (1996) suggested suggested that L. erythropterus ，Bangana behri and Cirrhinus microlepis were probably moving downstream at the beginning of the rainy season ， but they were unsure ，組d presented presented an alternative hypothesis 出at L. erythropterus may be moving bo 出 upriver and downriver downriver in order to reach suitable spawn 泊g environments within the Khong District area. area. ROBERTS (1 993) also reported that L. erythropterus migrates downstream at 血e beginning beginning of the rainy season. Whil e we agree 血at C. microlepis probably moves downstream at at the onset of the rainy season ，we believe 白紙 L. erythopterus and B. behri move upriver at at that time of ye 釘， although we are not sure where they move later in the rainy season. The first author observed L. erythopterus and B. behri unsuccessfully attempting to jump over 出e Somphamit Waterfalls in May ，indicating that they were trying to move upriver. ROBERTS & BAIRD (1995) also suggested 出at L. erythopterus migrations at the beginning of of the rainy season 訂 e upriver. Like Like SINGHANOUVONG ET AL. (1996) ，we recorded a number of Probarbus jullieni (p a eun eun ta deng) fmgerlings weighing about 5 to 10 g each in wing trap catches. However ， only a few large individuals of this cyprinid carp were recorded. No migratory behavior during

this this se 拙 on has been noted for the species (BAIRD ET AL. ，1999A; RAINBOTH ， 1996). SINGHANOUVONG ET AL. (1 996) reported that various silurid and bagrid catfishes such as as Ompok spp. ，Kryptopterus spp. ，Hemisilurus mekongensis ，Belodontichthys dinema and Micronema spp. ， probably migrate up the Kh one Falls at the beginning of the rainy season. ROBERTS (1993) had suggested that Hemisilurus mekongensis ，Micronema spp. ，and Belodontichthys Belodontichthys dinema were upriver migrators in the Khone Falls at the beginning of the ramy season. We found very few Ompok bimaculatus in li catches ，and Kryptopterus spp. was not abundant abundant either. Li catches at Nok Kasoum provided us with little information about these species. species. However ，most silurid and bagrid catfishes are more active at night ，when they are all all primarily caugh t. Th e silurid catfish Micronema micronema and Micronema apogon made up only 0.1 % of of catches for the four years ，and it seems unlikely that either undertakes long distance migrations migrations (BAIRD ET AL. ，1999A). 百le same goes for Micronema bleekeri ，which made up MEKONG RIVER PANGASIDAE CATFISH MIGRATIONS 97 just just 0.07% of catches. However ，Micronema micronema is known to migrate up medium to to 1紅 ge streams to spawn at the beginning of the rainy season (BAIRD ET AL. ， 1999A). An other silurid catfish Hemisilurus mekongensis (p a nang deng) made up 0.5% of 出e catches catches for 出e four years. Th is species spends most of the low-water season in deep-water pools pools in the Mekong River (BAIRD ET AL. ，1999A) ， but may migrate out of those areas at the the beginning of the rainy season. However ，this primarily noctumal species is unlikely to migrate migrate long distances (BAIRD ET AL. ， 1999A). Nevertheless ，S町 GHANOUVONG ET AL. (1 996) reported reported that many H. mekongensis migrated up Hou Som Gnai channel of the Kh one Falls during during the rainy season. 百le silurid ca 出 sh BelodontichthY$ dinema (p a khop) may migrate up and down the Khone Falls ， but it is unlikely to move long distances (BAIRD ET AL. ， 1999A). It made up 0.4 3% of li catches. SINGHANOUVONG ET AL. (1996) suggested 血at it conducts upriver migrations migrations through the Kh one Falls during the rainy season. In December 1994 ，villagers in 出e Kh one Falls 紅 ea were surprised to catch ， for the first tirne in memory ，a number of of individuals of the species over a short period in ‘chip' cylindrlcal 住'aps ぬrgeting schools of of migrating Scaphognathops bandanensis below the Kh one Falls (ROBERTS & BAIRD ， 1995). 1995). From the four years of catch data available ，it has not been possible to detect any significant significant trends in catch declines or increases for the li fishery at Nok Kasoum Island. Th e largest catches were recorded 加 1994 ， the lowest in 1995 ，and moderate catches were documented in 1998 and 1999. However ， since only a set of two traps at one fishing site were monitored ，albeit one of the most important single wing trap sites in the Kh one Falls area ，it is likely that our fish catch data are not representative of the overall wing trap fishery fishery (BAIRD ， 1998).

Like Like B AlR D ET AL. (2003) ，we compared daily total fish catches for each ye 紅 against lunar lunar phases. However ， unlike the catches for the fence-filter trap fishery in the Nok Kasoum Island area ，which 紅 'e closely associated with lunar periods (BAIRD ET AL. ，2003) ， no association be 何回n lunar phases and peak catch periods for the wing trap fishery was found. found. Fig. Fig. 3 compares water discharge in the mainstream Mekong River at Pakse (approx 加lately 130 km upriver from the Khone Falls) and weekly total catches the for two wing traps at Nok Kaso un1 Island. Th e data indicate 白紙 peak wing 住ap catches occur shortly shortly after water levels begin to rise from their lowest points in the ye 訂. Fig. 4 compares water water discharge and total Pangasius conchophilus catches. 百le results 訂 'e similar to those shown in Fig. 3. Fig.5 Fig.5 comp 紅白 the daily total catches for the two wing traps monitored in 1994 and daily daily total catches of Pangasius conchophilus ， Pangasius bocourti ， Pangasius krempft and Pangasius Pangasius larnaudii. L 訂 ge quantities of P. conchophilus were caught on May 29 ， while relatively relatively small 創 nounts were caught the days before and a負er the peak period. fishing Fig. Fig. 6 presents 血e same type of data included in Fig. 5， except for 1995. Unlike 1994 ， there there were no major catch periods for Pangasius conchophilus during the 1995 wing 釘'ap fishing fishing season at Nok Kasoum Island. However ，wing traps situated in other channels in 出e Kh one Falls area caught large amounts of P. conchophilus in 1995. 1998 1998 and 1999 data 紅 'e presented in Figs. 7 and 8. U n1 ike 1995 ， but similar to 1994 ， large large quan 出ies of Pangasius conchophilus were caught in 1998 over 出e two nights of June June 13 and 14. 98 I AN G. B AIRD, M ARKS. F LAHERTY AND B OUN PH ENG PHYLAVANH

500,000 1,400,000

-Weight - Oiacharge 1,200,000

400.000

1,000,000

I 300.000 800,000 ! § :;:t f~.ooo 600,000

400,000

100,000

200,000

Figure 3. Total 1994 to 1999 fish catches for the wing trap {li) fishery at Nok Kasoum Island, Khong Di strict, Champasak Province, Lao PDR against Mekong Ri ver water di scharge

500.000 1,200,000

- P. conchophtl us 1,000,000 <100 ,000 - Discharge

800,000 1300,000 § ! 600,000 H ,_~ f 200.000 0 .<\00.000

100,000 200,000

Figure 4. Total 1994 to 1999 Pangasius conchophilus catches fo r the wing trap (li) fishery at Nok Kasoum Island, Khong District, Champasak Province, Lao PDR against mainstream Mekong River water discharge MEKONG RIVER PANGASIDAE CATFISH MIGRATIONS 99

947,960

500,000

1!1 P. larnaudii 400,000 DP. krempfi ill P. conchophi/us !ZIP. bocourti

300,000 § ~ iii ~ 200,000

100,000

0 ·············r-··- ·· · >. >. >. >. >. c c c c "5 "5 "5 "5 ro ro ro ro ro :l :l :l :l --, --, --, --, "5 ;:!; --, --, --, --, 7 ~ ~ .;, .,:, cb J, 0 ~ .,J- ;;:; ~ ~ gj cb N N N ""

Figure 5. 1994 total daily fish catches and total catches of selected Pangasius species for the wing trap (li) fi shery at Nok Kasoum Island, Khong District, Champasak Province, Lao PDR

500,000

I!I P lamaudii 400,000 DP. krempfi ill P. conclwphi/us !ZIP. bocourti

300,000 § E Cl 'iii ~ 200,000

100,000

0 ~ ~ I >. >. >. >. >. c c c c "5 "5 "5 "5 "5 ro ro ro ro ro :l :l :l :l- --, --, --, ;:!; ;:!; ;:!; --, 7 --, --, 7 ~ - 0 ~ .,J- ~ .;, J, ~ ro .;, ~ N cb N (") ~ N N

Figure 6. 1995 total daily fi sh catches and total catches of selected Pangasius species for the wing trap (/i) fishery at Nok Kasoum Island, Khong Di strict, Champasak Province, Lao PDR 100 I AN G. BAIRD, MARKS. FLAHERTY AND BOUNPHENG PHYLAYANH

599,800 500,000

!!1 P. lamaudii 400,000 DP. krempn Ill P. canc!Jap!Jilus OP. bacaurti

300,000 § E ·a;Cl 3: 200,000

100,000

0 » » » » » c c c Ill c :; :; :; :; :; Ill ;,;; ;,;;Ill Ill ;,;;Ill --,::J --,::J --,::J --,::J --, --, --, --, 7 ~ ~ .;, d, ,;, 0 ..t ~ rJ, .;, d, N cD ~ N ~ gj N ~ N "'

Figure 7. 1998 total dail y fish catches and total catches of selected Pangasius species for the wing trap (li) fishery at Nok Kasoum Island, Khong District, Champasak Province, Lao PDR

500,000

99,300

l!lP. Iamaudii 400,000 DP. krempn Ill P. canchaphilus 12lP. bacaurti

300,000 § £, ·a; 3: 200,000

100,000

0 » » » c c c :; :; :; :; :; » » c --, --, ...., Ill Ill Ill Ill Ill ::J ::J --,::J --,::J --, 7 ;,;; --, 7 ,;, 0 ..t ~ ~ ~ .;, N a, cD N ~ rJ, ~ en N gj N "'

Figure 8. 1999 total dail y fish catches and total catches of selected Pangasius species fo r the wi ng trap (li) fishery at Nok Kasoum Island , Khong District, Champasak Province, Lao PDR MEKONG RIVER PANGASIDAE CAτ 下ISH MIGRA TIONS 101

In In 1999 ，peak Pangasius conchophilus catches again occurred over a short period.τ 'h e largest largest quantities were landed on May 14 ， and there were moderate catches on May 15. As 泊 1994 ，few P. conchophilus were caughtjust before or immediately after the peak fishing periods periods in 1998 and 1999.

DISCUSSION

百lere is not a close association between early rainy season pang 部 ud migrations and lunar lunar phases. Fishers in 血.e Kh one Falls 釘 ea have long maintained 白紙 changes 泊 wa 飽r levels 釘 e the main factor affecting fish migrations in the early rainy season (BAIRD & F'LAHE RTY ，2004A; BAIRDET AL. ， 2001B; BAIRDET AL. ， 1999A; BAIRD ， 1998; ROBERTS & BAIRD ， 1995; ROBERTS ， 1993). However ， with only four years of water discharge data available available for comparison with the wing trap fish catches ，it is difficult to confrrm 血e extent to to which hydrological factors 紅 'e influencing fish migrations up the mainstream Mekong River. River. S別 GHANOUVONG ET AL. (1996) reported that changes in tempera 伽 re and turbidity do not appe 釘 to greatly a町ect migrations at the beginning of rainy se 踊 on at the Kh one Falls. Falls. We believe ，出 they do ，出 at a combination of changes of water levels and currents 釘 'e probably the main factors affecting migrations. However ，it is unclear whether absolute wat 町 levels 釘 'e the key ， or if relative changes in water levels are more importan t. Local fishers fishers at Nok Kasoum state that large migrations of Pangasius conchophilus migrate upriver upriver when water levels reach a certain height ， which fishers mentally mark with the location location of the roots of certain seasonally inundated trees near 白eir traps. We have also observed 白紙 large quan 出ies of P. conchophilus are typically caught shortly after water flow flow volumes suddenly increase ， as have SINGHANOUVONG ET AL. (1 996). More analysis of already already collected data and additional research will be required befo 問 the fish migrations C釦 be identified and understood. P. conchophilus may in fact only migrate relatively short distances. distances. Th ey have been one of the main reported beneficiaries of Fish Conservation Zones established by villagers to protect fish 白at congregate 泊 deep-water areas (BAIRD ET AL. ，2004B) It It is difficult to know 白 e directionality of the migrations of different fish species found found in li catches ， but we have hypothesized above about the directionality of various species. species. Our conclusions 訂 e based on documented biology and ecology ，ye 釘 s of field observations observations from the Kh one Falls area ， and local ecological knowledge provided by villagers villagers from many communities in Khong Dis 凶ct since 1993 (see BAIRD & FLAHERTY ， 2004B; BAIRD ET AL. ， 1999A; ROBERTS & BAIRD ， 1995; ROBERTS ， 1993). Combining scientific scientific data collection methods ，field observations and local experience is an invaluable means of learn 泊g about 出e migration patterns of Mekong River fish ， as local people from southern southern Laos generally have extensive experience and knowledge regarding Mekong Ri ver ecology ecology and fish biology (BAIRD &孔 AHERTY ， 2004B; BAIRD ET AL. ， 1999A). τ'h e occu 町'ence of relatively 1紅 gequ 組 tities of Pangasius hypophthalmus in li catches in in 1995 ， with virtually no individuals be 泊g landed during the other three years data were collected ， provides some important clues potentially useful for the management of juvenile P. P. hypophthalmus in Cambodia. BARDACH (1959) repor 凶伽t P. hypophthalmus fingerlings 釘 e caught south of Phnom Penh over a short period in June and July ，and sometimes in in August ， to supply cage and pond culture operations in Vietnam ， and Cambodia to a 102 102 弘N O. BAIRD ，MARK S. FLAHERTY AND BOUNPHENG PHYLAVA 聞

lesser lesser extent in Cambodia (NOOR. 1999). NOOR (1999) rep 'O rted 血at in 1994 the Camb 'O dian g'O vernment banned the c'O llecti 'O n 'O f juve 凶le P. hypophthalmus. which are c 'O llected ne 紅 Phn 'O m Penh ，and 加 Kandal and Kamp 'O ng Cham pr 'O vinces ，Centr a1 Camb 'O dia (NOOR ， 1999) ， using bag nets made 'O fm 'O squit 'O netting. H 'O wever ， the ban has apparently n'O t been enf 'O rced and has been largely ineffective (Fisheries Offici a1 s， Department 'O f Fisheries ， pers. pers. c'O mm. ， 2000). Acc 'O rding t 'O Camb 'O dian fisheries 'O ffici a1 s. h'O wever ，a number 'O f junks c釘可泊g quantities quantities 'O f live juvenile P. hypophthalmus t'O Vietnam were sunk 町 damaged in August 1994 1994 near Neak Lu 'O ng ，resulting in the release 'O f large numbers 'O f fish int 'O the l'O wer Mek 'O ng River. NOOR (1 999) was unsure 'O f the bi 'O I'O gic a1 impact 'O f banning the fishery. H 'O wever ， the 'O ccurrence 'O f significant numbers 'O f 0+ year class P. hypopthalmus in 1995 li li catches indicates 白紙 the release 'O f large numbers 'O f juve 凶les int 'O the Mek 'O ng River 泊 1994 (b 田 ic a1 1y the equiv a1 ent 'O f reducing the fishing eff 'O rt 'O n 血 e juveniles) may have resulted resulted in m 'O re larger fish migrating up t'O出 e Kh'O ne Fa1 1s during the f'O ll 'O wing ye 低官le lack lack 'O f P. hypophthalmus in li catches d町泊g 1998 and 1999 a1 s 'O fits with this hyp 'O thesis ， since since n 'O barges carrying fish have app 訂 'ently been sunk since 1994. BARDACH (1 959) suggested that capturing fry and fingerlings 'O f Pangasius hypophthalmus hypophthalmus (listed by him as Pangasius sutch i)， P. larnaudii and Pangasius sp. (listed as as Pangasius micronema) was n 'O t destructive because their c'O llecti 'O n actually kept fish a1 ive that would have died 'O f natur a1 m 'O rt a1 ity in the wild anyway. He expected that natur a1 m 'O rtality f'O r 合y at the stage 泊 life 白紙 they were being c 'O llected was 90 t 'O 95%. BARDACH (1 959) a1 s 'O rep 'O rted 白紙P. hypopht 加，lmus migrate upriver from cen 回 lCambodia to 白 en 'O rtheast Camb 'O dian pr 'O vince 'O fS 知ng Treng and ont 'O白 e Kh'O ne Fa1 1s at 出 e beginning beginning 'O f the rainy seas 'O n， where they spawn. As described ab 'O ve ，HOOAN EI AL. (2004) (2004) have recently pr 'O vided s 'O lid evidence that these migrati 'O ns 'O ccur. H 'O wever ，we kn 'O w little ab 'O ut spawr 血 g activities f'O r this species ar 'O und the Kh'O ne F a1 1s. M 'O re research sh 'O uld be dev 'O ted t 'O this imp 'O rtant regi 'O nal management questi 'O ns related t 'O P. P. hypophthalmus. Sever a1 meth 'O d'O I'O gic a1 issues 紅'O se during 出 is s制 dy. At the outset ，sever a1 decisiohs were made regar ・ding the most appropriate s住ategy f'O r data collection. Collect 泊g too much data data is a waste of limited resources ， while collecting too few data can leave researchers with with anin 組問凶，te basis for 組 swering questions 加 p 'O rtant for making management decisions (COWX ，1995). 官邸data collection program was 1釘 'gely exploratory 泊 nature ，and sampling was theref 'O re done on a basis daily ，rather 白血 on days randomly selected or spaced over some interv a1 of time. In con 回 st ，S町 OHANOUVONO ET AL. (1 996) decided to 蜘 nple the li li fishery at Hou Som Gn ai 血ree days a week. Whi le staggered fishery sampling is very cost-effective ，0 町 investigation of the li fishery suggests 白紙 daily sampling is preferable owing to the nature of the fishery. 百le bulk of the fish caught in the li fishery 釘 'e often landed landed during a sin 副e night ， or over tw 'O nights. For example ，on 28 May 1994 ，950 g of Pangasius Pangasius conchophilus were caught at the traps at Nok Kasoum ， but on 29 May 1994 ，血 e catch catch increased almost 1 ，000-fold t 'O 947 ，960 g. 百le catches of P. conchophilus were just 26 ， 100 g 'O n 30 May 1994. Had our da 飽 collecti 'O n been spaced at some interval 'O f time ， it it is quite possible 白紙 we would have missed the main run of P. conchophilus. This ，泊 turn ，would have resulted in a significant under estimation of catches. M EKONG R IVER PANGAS ID AE CAT F IS I-I M IG R AT IONS 103

Fi glll 官 9. Th 巴 thircl alll hor acl cl s a flln nel b askel 10 on e of Ih e IWO li Ir ap s al Nok Ka SO ll I1l Isla ncl ‘ jll sl b巴low Ih 巴 K l1 0n 巴 Fa ll s. I1l 0n il O recl c1 l1 ring Ih iss llI cl y. T I1 巴se ba skels 山 巴 onl y aclcl ecl w l1 en la rg e (ドtanli lies of /-I ell ico rh YIl c/ lII S lo !J O(IIS are being ca ll ghl c1l1 rin g Ih e c1 ayli I1l e

Fi gul 巴 10. On e of th 巴 I1l an y Ii Ir ap s lI se cl in Ih e K hone Fa ll s area 01. Ih 巴 m aIn Slr ea m M 巴kong R ive l 川 sOlllh 巴rn L aos 同 一- E- 司 口 =凶 町U ε FZ ・5 ニ -需主 u ppmD - 的 D 匂「 百戸- E 司出U E。 P= 一 Z 世 Z〈〉〈」〉 〉J 』一 回 司- 三- = 戸- o =ι 002戸高』a J -E 07 nu 心 --コ E U Zコ 司， 古 DDE』D 己 O 勾 凶 ω 百三芝 CZ〈〉ト -コ ロ己戸 山叫 己 己凶工 E- = m 〈 J ~~ F主 h - u 一叫 主ZU m 主 出 的 N2E 信 三己主古 一u ココ ON-， 』D}F c U -〈 u -24- - ∞.0 = 元三 -wu z - D戸 ω - 一 沼 -元-=喝比 Z Z〈 ω 間 -』二 一 U 。ωh句 c』回目UUE - C u Z 呂 z リ品 ∞ D 己 戸一 ω AC--一一一Z H Z -

回 -コ 寸。 叩 - n一 MEKONG RIYER PANGAS IDAE CATF ISH M IGRAT IONS 105

Fi gllr e 13. Cypl 川 id s co ll 巴clin g in a li lr ap al Nok Ka so llm Islancl clu rin g lh 巴 clayl im e. Th 巴 Pan gas iicl ca lfi sh，lh 巴 ma ins lay of lh巴 li sh巴ry，are ma inly cι 11I ghl al ni ghl

Fig山 巴 14. A wo m an co ll ec lss l11 all fis h from a li lr ap al Nok Ka soll l11 Is lancl. A II sli cle s by lan G. Ba ircl 106 106 IAN G. BAIRD ，MARK S. FLAHERTY AND BOUNPHENG PHYLAVANH

CONCLUSIONS

This This study was undertaken to better understand the wing trap fishery in the lower reaches reaches of the Mekong River ，and to investigate whether a relationship exists between pangasid pangasid migrations and lunar cycles and/or hydrological cycles in the Mekong. There is some evidence that water discharge levels and associated changes in cuη'ent regimes are the the primary triggers for m 組 Y of the bi-directional migrating that fish have been documented from the li fishery at the Kh one Falls. Yet ，much is still unknown about the biology and life life cycles of the many species caught in li traps. Current information on the main characteristics characteristics of Mekong mainstream migrations is highly fragmented and often only qualitative qualitative in origin. The fact that the taxonomy of many Mekong fish species is confused adds adds further to difficulties in the documentation process (ROBERTS ， 1997; SlN GHANOUVONG ET AL. ， 1996). τ'h e pangasiid catfishes ，including Pangasius conchophilus ， P. larnaudii ， P. bocourti and P. krempfi are the main highly migratory species t紅 geted by this fishery at Nok Kasoum Island ， but highly migratory minnows and ca 叩 s such as Henicorhynchus lobatus ， Henicorhynchus siamensis ，Labiobarbus leptocheilus ，Loboche i/ us melanotaenia ， Crossocheilus Crossocheilus reticulatus ，Paralaubuca り'pus and Scaphognathops bandanensis are also caught caught in large quantities. Other cyprinid carps ， such as Cosmocheilus harmandi ， Cyclocheilichthys Cyclocheilichthys enoplos ，Puntioplites falc 俳r and Hypsib α rbus malcolmi ，make up significant significant proportions of the catch. Bagrid ，silurid and sisorid catfishes like Hemibagrus nemurus ，Hemibagrus wyckioides ， Bagarius yarrelli ，Micronema spp. and Hem η1i ωおsilu ωu川1η 'u 凶tβ S mη1ekongen 幻“ si C凶at 凶f白ishe 邸sa訂re unlikely to be long distance mi 抱gratωor 凶s，and may only disperse short distances at at the beginning of the rainy season. Considering Considering the importance of the wing trap and other fisheries in the Khone Falls area area to the welfare of local subsistence communities ，it is critical that wise management decisions decisions be made at local ，national ， and regional levels. Without considering various scales ，it will be difficult to achieve effective managemen t. In the face of growing development development pressures ，many of which promise short-term economic gains ，it is essential to to ensure that this important resource is protected and managed for the long-term. While there there are many cn に ial management issues to consider in relation to the fish populations that that the li fishery depends upon ， the construction of large hydroelectric dams and water diversion diversion projects on the mainstream Mekong River and her large tributaries are the most worrying. worrying. Dams could block important fish migrations ，and fundamentally alter the ecology of of rivers near the dam ，and also downstream areas affected by changes in hydrology caused by dam construction and operation. There are many planned dams in the Mekong basin that are viewed with grave concem by the intemational environmental community (HIRSCH & WY Aπ ， 2004; BAIRD ET AL. ， 2003; 2003; BAIRD ET AL. ，200 1B; INTERNA Tl ONAL RIVERS NETWORK ， 1999; CLARIDGE ET AL. ， 1997; 1997; MCCULLY ， 1996; ROBERTS & BAIRD ， 1995). However ， the ten dams formerly envisioned envisioned by the Mekong River Commission for the mainstream Mekong River (MEKONG SECRETARIAT ，1994) ， as well as dams planned in the Sekong ，Se San and Sre Pok basins in in Viet Nam ，Cambodia and Laos (HALCROW & PARTNERS ，1998) ， are of greatest concem to to the future welfare of fishing-dependent rural communities in Laos. Already ， the Yali Falls Falls dam on the Se San River in Viet Nam has caused dramatic changes in river hydrology MEKONG RIVER PANGASIDAE CATFISH MIGRATIONS 107

and and water quality that have seriously negatively impacted riverine habitat and associated fisheries fisheries in the Cambodian provinces of Ratanakiri and Stung Treng (HIRSCH & WYATT ， 2004; 2004; BAIRD & FLA 田 RTY ，2004A; BAIRD ET AL. ， 2002). Artisanal fishing provides the backbone backbone for local food security and contributes heavily to the vilI age economies in communities communities scattered throughout the reaches of the Mekong River and its tributaries. Large Large dams constructed on the mainstream Mekong River or her larger tributaries ，including in in China ， could devastate fisheries in a number of countries in the basin (BAIRD & FLAHERTY ， 2004A; BAIRD ET AL. ， 2003; BAIRD ET AL. ， 2001B; BAIRD ET AL. ， 1999A; ROBERTS & BAIRD ， 1995). With so many people's Ii velihoods at stake ， great caution is required before proceeding proceeding with these projects.

ACKNOWLEDGMENTS

We would like to acknowledge Mr， Phongsavath Ki souvann aI ath ， Mr. Vixay In thaphaisy ， Mr. Mr. Bounhong Mounsouphom ， Mr. Soukan Phimthong ， Mr. Khamsouk Xaiyamanivong ， Mr. Mr. Bounthong Vongsenesouk ， Mr. Zeb Hogan and Mr. Peter Cunningham for their contributions contributions in collecting and compiling fisheries data from Ban Hang Kh one at various times times between 1993 and 1999. Thanks to Dr. Warren Brockelman and another 加 onymous reviewer reviewer for refereeing this article. M r. To Mounsouphom entered most of the data into 出e computer computer database with the assistance of the Vientiane-based NGO CIDSE. Mr. Pierre Dubeau ，a former CUSO co-operant ， helped develop the Ban Hang Khone fisheries database. Ms. Sarah Adams extracted data from the database to create the graphs and tables included in 出is report. Dr. Tyson R. Roberts provided valuable input regarding the data collection program at Ban Hang Kh one during the early stages ，and helped identify many of the fish species species coIIected at Ban Hang Khone. Mr. Ole Heggen from the Department of Geography of of the University of Victoria in Canada prepared the maps. Research conducted in 1994 and and 1995 was done under the auspices of the Lao Community Fisheries and Dolphin Protection Protection Project (LCFDPP) ，and work completed in 1998 and 1999 was conducted for the the Environrnent aI Pr otection and Community Development in Siphandone Wetland Project (EPCDSWP) ， implemented by the NGO CESVI Cooperation and Development and funded by the European Commission. We would like to thank the Agriculture and Fore 柑 y Di vision of of Champasak Pr ovince ， the Agriculture and Forestry Office of Khong District ，and the Center Center for Protected Ar eas and Watershed Management (CPAWM) at the Department of Forestry ，Vientiane ， for their suppo 口. Speci aI thanks to Mr. Vilay Athn 詑 hone ， Mr. Nouphay Athn 北hone ， Mr. Sounthone and Mrs. Len Athn 北hone ，Mr. Kiheng ，Mr. Souk ，M r. Yeun ， Mr. Mr. Say and Mr. Souan for aIIowing us to examine their wing trap fish catches at the southem end ofNok Kasoum Island in 1994 ，1995 ，1998 釦 d 1999 ，and for explaining their local local ecologic aI knowledge regarding their fishing activities to us. Without their patience and understanding ，出 is research would not have been possible. 108 108 IAN G. BAIRD ，MARK S. FLAHERTY AND BOUNPHENG PHYLAVANH

REFERENCES

BAIRD ，1. G. 1998. Pr e1iminary Fishery Stock Assessment Resu1ts from Ban Hang Khone ，Khong Di strict ， Champasak Province ，Southem Lao PD R. Technic a1 report prepared for the Environmental Protection and Community Development in Siphandone Wetland pr 吋ect ，CESVI ，Pakse ，Lao PDR ， 112 pp. BAIRD ，1. G. 200 1. Aquatic biodiversity in the Siphandone wetlands. Pages 61-74 in: G. Daconto (ed.) ，Siphandone Wetlands ，CESVI ，Bergamo ，Italy ， 192 pp. BAIRD ，1. G. ， V. INTHAPHAISY ， B. PHYLAVANH ，AND P. KISOUVANNALATH. 1998. A Rapid Fisheries Survey in Khong Dis 凶ct ，Champasak Province ，Southem Lao PDR. Technical repo 目 prepared for the Environmen-

tal tal Protection and Community Development in Siph 組 done Wetland Pr 吋ect ，CESVI ，Pakse ，Lao PDR ， 31 31 pp. BAIRD ，1. G. AND B. PHYLAVANH. 1999. Fishes and Forests: Fish Foods and the Importance of Seasonally Flooded

Riverine Riverine Habitats for Mekong River Fish. Technical report p陀 pared for the Environment a1 Protection and Community Deve10pment in Siphandone Wetland Project ，CESVI ，Pakse ，Lao PDR ，46 pp. BAIRD ，1. G. ，V. INTHAPHAISY ， P. KISOUVANNALATH ， B. PHYLAvANH ，AND B. MOUNSOUPHOM. 1999a. Fishes 01 Southern Southern Laos. Lao Community Fisheries and Dolphin Protection Projec t. Ministry of Agriculture and Forestry ，Lao PDR ， 161 pp. (in Lao) ， BAIRD ，1. G. ，V. INTHAPHAISY ， P. KISOUVANNALATH ，B.'VONGSENESOUK ，AND B. PHYLAVANH. 1999b. Th e Setting Up and the In iti a1 Resu It s of a Villager Based System for Monitoring Fish Conservation Zones i目白e Mekong River ，Khong Di strict ，Champasak Pr ovince ，Lao PD R. Technical report prepared for the Environmental Environmental Protection and Community Deve10pment in Siphandone Wetland Project ，CESVI ，Pakse ， Lao PDR ，41 pp +. BAIRD ，1. G. ， B. PHYLAVANH ， B. VONGSENESOUK ，AND K. XAIYAMANIVONG. 2001a. The ecology and conservation of of the smallsc a1 e croaker Boesemania microlepis (Bleeker 1858-59) in the mainstream Mekong River ， southem Laos. Nat. Bull. Hist. Siam Soc. 49: 161-176. BAIRD ，1. G. ， Z. HOGAN ， B. PHYLAIVANH ，AND P. MOYLE. 2001b. A communal fishery for the migratory ca 凶 sh Pangasius Pangasius macronema in the Mekong River. Asian Fisheries Science 14: 25 -41. BAIRD ，1. G. ，M. BAIRD ，CHUM MONI CHEATH ，KIM SANGHA ，NUON MEKRADEE ，PHAT SOUNITH ，PHOUY BUN NYOK ，

PROM SARIM ，Ros SAVDEE (PHIAP) ， H. RUSH 苧ON ，AND SIA PHEN. 2002. A Community-Based Study of the Downstream Im pacts of the Yali Falls Dam A10ng the Se San ， Sre Pok and Sekong Rivers in Stung Treng Province ，North 巴ast Cambodia. Se San Protection Network Project ，Partners For Deve10pment (PFD) ，Non Timber Forest Products Project (NTFP) ，Se San District Agriculture ，Fisheries and Fores 町 Office ，and Stung Treng Di strict Office ， Stung Treng ，Cambodia. BAIRD ，1. G. ，M. S. FLAHERTY ，AND B. PHYLAVANH. 2003. Rhythms of the river: lunar phases and migrations of small small carps (Cyprinidae) in the Mekong River. Nat. Bull. Hist. Siam Soc. 51 (1): 5-36. BAIRD ，1. G. ，AND M. S. FLAHERTY. 2004 A. Beyond borders: national important Mekong River medium sized migratory migratory carps (Cyprinidae) and fisheries in Laos and Cambodia. Asian Fisheries Science. 17 (3 -4): 279-298. BAIRD ，1. G. ，AND M. S. FLAHERTY. 2004B (i n press). Mekong River fish conservation zones in southem Laos: assessing assessing e百'ectiveness using local ecological know1edge. Environmental Management. BARDACH ，J. E. 1959. Report on Fisheries in Cambodia. USOM ，Cambodia ，Phnom Penh ，55 pp. BERNATCHEZ ，1.， AND J. DODSON. 1987. Relationships between bioenergenetics and behaviour in anadromous fish fish migrations. Canadian Journal 01 Fisheries and Aquatic Sciences 44: 399 -4 07. BROCKELMAN W. Y. 2002. Editorial: Riverine Natural History. Nat. Hist. Bull. Siam Soc. 50(1): 1-2.

CLARIDGE ， G. F. ， T. SORANGKHOUN ，AND 1. G. BAIRD. 1997. Com 澗 unity Fisheries in Lao PDR: A Survey 01 Techniques Techniques and Issues. IUCN-The World Conservation Union ，Vientiane ，Lao PD R. Cowx ，1. G. 1995. Fish Stock Assessment-A biological basis for sound ecological managemen t. Pages 375-388 in: in: D. M. HARPER AND A. J. D. FERGUSON (eds.) ， The Ecological Basislor River Management. John W iI ey & Sons Ltd. ，Chichester ，UK. DACONTO ， G.(ED.) 200 1. Siphandone Wetlands ，CESVI ，Bergamo ，Italy ， 192 pp. DEAP ，L.， P. DEGEN ，AND N. VAN ZALINGE. 2003. Fishing Gears 01 the Cambodian Mekong. In land Fisheries Research Research and Development Institute of Cambodia (IFReDI) ，Phnom Penh ，Cambodia ，269 pp.

Food 飢 d Agricul 旬 re Organization (FAO). 1999. Fishery Count η ， Profiles. FAO ，Rome ，Italy. Halcrow & Partners ， Ltd. 1998. Interim Environmental Examination. Sekon g--S esan and Nam Theun Rivel Basins Basins Hydropower Study. Asian Development B 釘11<， Manila ，Philippines. MEKONG RIVER PANGASIDAE CATFISH MIGRATIONS 109

HIRSCH ，P. ，AND A. W 刊行， 2004. Negotiating locallivelihoods: Scale of conflict in the Se San River Basin. Asia Pacific Pacific Viewpoint 45(1): 51 -6 8. HOGAN ， Z. S. ， P. B. MOYLE ， B. MAY ，M. 1. VANDER ZANDEN ，AND l. G. BAIRD. 2004. The imperiled giants of the the Mekong. American Scientist 92: 228-237. INTERNATIONAL RIVERS NETWORK. 1999. Power Struggle. The lmpacts o[ Hydro-Development in Laos. In tema- tion aI Rivers Network ，Berke1ey ，CA ，USA ，68 pp. LENORMAND ， S. 1996. Les Pangasiidae du delta du Mekong (Viet Nam) ・description preliminair 巴 des pecheries ， elemets elemets de bi10gie et perspectives pour une diversification des elevages. Memoire ENSAR ，Orstom-Gamet ， France. France. LIENG ，S. ， C. YIM ，AND N. P. VAN ZALINGE. 1995. Fisheries of the Tonlesap River Cambodia ，1: The Bagnet (Dai) Fishery ・Asian Fisheries Science 8: 258-265. 品1CCULLY ， P. 1996. Silenced Rivers: The Ecology and Po /i tics o[ Large Dams. Zed Books ，London ，UK. MEKONG SECRETARIAT 1994. Mekong Mainstream Run-of-River Hydropower. Executive Summary ，Bangkok ， Th ailand ，20 pp. NGOR ， B. P. 1999. Catfish flηcollection in the Mekong River of Kandal fP hnom Penh. Pages 12 4- 129 in N. P. VAN ZALINGE ， T. NAO ，AND L. DEAP (eds.) ， Present Status o[Cambodia's Freshwater Capture Fisheries and Management lmplications. Nine presentations given at the Annual Meeting of 出e Dep 紅佃lent of Fisheries Fisheries of the Ministry of Agriculture ， Forestry and Fisheries ， 19-21 January 1999. Mekong River Commission and Department of Fisheries ，Phnom Penh ，Cambodia. NORTHCOTE ， T. G. 1984. Mechanisms of fish migrations in rivers. ln J. D. MCCLEAVE ， G. P. ARNOLD ，1. J. DODSON ，AND W. H. NEILL (eds.) ，Mechanisms o[ Migrations in Fishes. Plenum Press ，New York ，574 pp. RAINBOτH ，W. 1. 1996. Field Guide to the Fishes o[ the Cambodian Mekong. Food and Agriculture Organization of of the United Nations ，Rome ，Italy ，265 pp +. ROBERTS ， T. R. 1993. Artisinal fisheries and fish ecology below the great waterfalls of the Mekong River in Southem Laos ， Nat. Hist. Bu l/. Siam Soc. 41 (1): 31 -6 2. ROBERTS ， T. R. 1997. Systematic revision of the tropical Asian labeoin cyprinid fish genus Cirrhinus ， with

descriptions descriptions of new species and biological observations on C. lobatus 目 Nat. Hist. Bu l/ 目 Siam Soc. 45: 171-203. 171-203. ROBERTS T. R. ，AND C. VIDTHAYANON. 199 1. Revision of the tropical Asian catfish family Pangasiidae with bio10gical bio10gical observations and descriptions of three new species. Proc. Philadelphia Acad. Nat. Sci. 143: 97-144. 97-144.

ROBERTS ， T. R.， AND T. J. WARREN. 1994. Observations on fishes and fisheries in Southem Laos and North 巴astem Cambodia ， October 1993-February 1994 ， Nat. Hist. Bu l/. Siam Soc. 42: 87-115.

ROBERTS ， T. R. ，AND l. G. BAIRD. 1995. Traditional fisheries 佃 d fish ecology on the Mekong River at Kh one Waterf aI ls in Southem Laos. Nat. Hist. Bu l/. Siam Soc. 43: 219-262 SINGHANOUVONG ，D. ， C. SOULIGNAVONG ， K. VONGHACHAK ， B. SAADSY ，AND T. 1. WARREN. 1996. The main wet- se ぉ on migration through Hoo Som Yai ， a steep-gradient channel at the great fault line on the Mekong River ，Champassack 針。 vince ，Southem Lao PDR ， Indigenous Fisheries Dev 巴lopment Pr oject ，Fisheries Eco10gy Technical Report ，Vientiane ，Lao PDR ， 4: 1-115. VAN ZALINGE ， N. P. ，NAO THUOK ，AND L. DEAP. (eds.) 1999. Present Status o[Cambodia's Freshwater Capture Fisheries Fisheries and Management lmp /i cations. Nine presentations given at the Annual Meeting of the Depart- ment of Fisheries of the Ministry of Agriculture ， Forestry and Fisheries ， 19-21 January 1999. Mekong River River Commission and Department of Fisheries ，Phnom Penh ，Cambodia ， 149 pp VAN ZALINGE ， N. P. ，NAO THUOK ，AND TOUCH SEANG TANA. 2000. Where there is water ， there is fish? Cambodian Fisheries Issues in a Mekong River Basin Perspective. Pages 37 -4 8 in: M. AHMED ，AND P. P. HIRSCH (eds.) ，Common property in the Mekong: issues o[ sustainability and subsistence. ICLARM Stud. Stud. Rev. 26 ，Man i1 a，Ph i1 ippines ，67 pp. WARREN ， T. J. ， G. C. CHAPMAN ，AND D. SINGHANOUVONG. 1998. Th e Upstream Dry-Season Migrations of Some Important Important Fish Species in the Lower Mekong River of Laos. Asian Fisheries Science 11: 239-25 1.