Effects of Severe Drought on Freshwater Mussel Assemblages

-We examined changes in freshwater mussel abundance and species composition at eight sites in Alabama and Mississippi in response to a severe drought in 2000. Five small-stream sites in Bankhead National Forest were heavily impacted by drought; one site dried almost completely, and four sites experienced total or near cessation of flow but retained water in their channels to a large extent. In contrast, three large-stream sites retained flow and experienced only minor streambed exposure, primarily along the stream margins. In small streams, overall mussel density before and after the drought declined by 65-83%, and the magnitude of the decline did not differ among streams regardless of whether the channel dried or remained wetted. Introduced Corbiculafluminea (Asian clam) experienced near total mortality and declined to a greater extent than native unionids. The magnitude of decline was similar among unionid species, and the likelihood of surviving the drought was mostly a function of predrought abundance; differences in drought tolerance among species were not evident. Consequently, assemblage composition changed primarily because of the loss of rare species, resulting in drainagewide homogenization and convergence on a shrinking species pool. In contrast, we found no evidence for changes in the total abundance or composition of mussel assemblages in large streams that continued to flow during the drought. Our results show that mussels are highly sensitive to the secondary effects of drought-most likely the low levels of dissolved oxygen caused by low flow, wann temperatures, and high biological oxygen demand-in addition to the direct drying of their habitat. The postdrought abundances of some species in Bankhead National Forest may now be below the minimum necessary for successful reproduction. These populations, which are isolated by reservoirs, may be in a downward spiral from which they will have difficulty recovering in the absence of immigration. Like many other endangered organisms, remnant populations of freshwater mussels are often small and isolated from other populations. Some species survive as reproducing populations only in a single, short stream reach (e.g., Rogers et al. 2001), and many others are represented by a few small, widely separated populations (Shelton 1997; Parmalee and Bogan 1998; Jones et al. 2004; Warren and Haag 2(05). Even in ostensibly well-protected conservation refuges, small refuge size and isolation render these populations highly vulnerable to extinction through loss of genetic variability, chance fluctuations in reproduction and survivorship, and environmental disturbance (Gilpin and Soule 1986). Drought is one of the most pervasive fonns of natural environmental disturbance in inland aquatic ecosystems, especially in small streams, but the generalized effects of drought on mussel communities are poorly understood. Previous workers considered freshwater mussels as a group well adapted to survive prolonged emersion (Byrne and McMahon 1994), but this generalization was based on results from a small number of wetland-adapted lowland species, which do * Corresponding author: [email protected] Received April 27, 2007; accepted December 20, 2007 Published online July 10, 2008 not accurately represent the majority of the North American fauna. One species, Uniomerus tetralasmus (pondhorn), can persist in ephemeral wetlands by burrowing into the substrate (Frierson 1903) and can survive emersion for nearly 2 years (Hol1and 1991). Other lowland species, including Toxolasma paulus (iridescent lilliput), T. texasiensis (Texas lilliput), and U. caroliniana (Horida pondhorn), can survive moderate periods of emersion or low levels of dissolved oxygen (Holland 1991; Gagnon et al. 2004). In contrast, many riverine mussels and the introduced Corbicula fluminea appear to be intolerant of low dissolved oxygen and emersion, especially in warm temperatures (Holland 1991; Bartsch et al. 2000; McMahon 2(01) and sustain high mortality when stranded in drying pools or on stream margins (Metcalf 1983). Mussels can move short distances to deeper water in response to receding water levels (Coker et al. 1921; White 1979). In large streams with pennanent flow and resultant high levels of dissolved oxygen, this response probably decreases drought-associated mortality (Golladay et al. 2004). In small, shallow streams, the rarity of deeper refuges coupled with higher likelihood of low flows and low dissolved oxygen renders many mussel species vulnerable to emersion and other drought effects (Gagnon et al. 2(04). Because mussel species occurring in different habitats may differ substantially in drought tolerance, context-