Sudden crash of a local population of the endangered bagrid catfish Pseudobagrus ichikawai

Abstract

Listen

Translate article

The bagrid catfish Pseudobagrus ichikawai is an endangered freshwater fish (EN in the Red Data Book of Japan; Ministry of the Environment 2003) and distributed only in a narrow region of central Honshu, Japan. The catfish inhabits the middle reaches of rivers in well-protected environments and thus is particularly susceptible to any introduced water pollution or human-made changes to habitat structure (Watanabe 1997; Kobayakawa 2003). Several studies have addressed the conservation (e.g., Watanabe and Mori 1998; Mori and Watanabe 1999) and basic ecology (e.g., Watanabe 1994, 2008; Tashiro et al. 2005) of this species. A long-term ([ten-year) study conducted in a tributary of the Nagara River system, Gifu Prefecture, indicated that the population of P. ichikawai fluctuates greatly, primarily because of changes in recruitment among years, likely to be because of regional or climatic factors such as rain-induced floods (Watanabe and Ichiyanagi 2006, unpublished data). Conservation efforts have focused on reducing the effects of humanmade changes to river environments while monitoring populations (e.g., Watanabe and Mori 1998). During the long-term study mentioned above, I observed a drastic decline in the population between 2007 and 2008 in a 1.2-km span of the river, about half the range of P. ichikawai. Here, I report the details of this sudden population crash and discuss its possible causes. The tributary of the Nagara River that was studied is about 17 km long, and P. ichikawai inhabits an area in its middle reaches of about 6 km, with a core region of about 3 km. Four small weirs (0.9–1.4 m in height) in the core area restrict upstream movement of most fishes, including P. ichikawai (Fig. 1). The catfish is patchily distributed in regions of slow flow, i.e., pools and glides (hereafter, ‘‘habitat units’’). The total number of catfish observed by night visual census (see Watanabe and Ito 1999) in summer (late July to early August) in 13 representative habitat units fluctuated between 263 and 1,019 (mean ± SD = 562 ± 267) from 2001 to 2007 (Fig. 2). In the lower and upper reaches of the third weir (1.1 m in height), located at the midpoint of the core section (W3; Fig. 1), 39–199 fish/ habitat unit (90 ± 54 for 7 habitat units; 633 total) and 12–45 fish/habitat unit (30 ± 12 for 6 habitat units; 179 total) were observed, respectively, on 30 July 2007. However, on 4–5 August 2008, only 0–36 fish/habitat unit (9 ± 13; 65 total) in the lower section and 39–63 fish/ habitat unit (48 ± 8; 289 total) in the upper section were observed. That is, the counts in all habitat units in the lower section dramatically decreased, while those in the upper section moderately increased from 2007 to 2008 (Figs. 1, 2). No such drastic decrease has been observed in this river at least since 1996 (Watanabe, unpublished data). In the lower section, except for a pool just below W3 (no. 7) inhabited by a usual population (36 fish observed) of various sizes, only a few small (\60 mm SL) fish were observed, mainly in the upper half of the area. These results observed on 4–5 August were roughly confirmed on 26 August 2008. Watanabe and Ito (1999) estimated the ratio of real population size to night-visual counts to be 4.27 ± 0.73 for this species. The population size in the lower section was thus roughly estimated at 2,700 (count = 633) in 2007 and at 280 (count = 65) in 2008; more than 2,400 individuals (ca. 90% of the 2007 population) disappeared between August 2007 and July 2008. These data indicate that this local population crashed K. Watanabe (&) Department of Zoology, Division of Biological Science, Graduate School of Science, Kyoto University, Kitashirakawa-Oiwakecho, Sakyo-ku, Kyoto 606-8502, Japan e-mail: watanak@terra.zool.kyoto-u.ac.jp