Fish kills, bottom-water hypoxia, and the toxic Pfiesteria complex in the Neuse River and Estuary

A recent paper by Paerl et al. 'Ecosystem responses to internal and watershed organic matter loading: consequences for hypoxia in the eutrophying Neuse River Estuary, North Carolina, USA' (1998; Mar Ecol Prog Ser 166:17-25) makes the statement that increased nitrogen loading to the Neuse River Estuary has led to algal blooms which produced organic matter loads capable of causing extensive hypoxic and anoxic conditions that, in turn, have induced widespread mortality of resident finand shellfish (p l?) . In this Comment we demonstrate that Paerl et al.'s central conclusion about finfish kills is not supported either by their data or by any statistical analysis, despite invoking predictability ('Results and discussion', p 20-24, Paerl et al. 1998). The paper contains numerous misinterpretations and misuse of literature citations. Paerl et al. also made serious errors of omission, germane from the perspective of science ethics, in failing to cite peerreviewed, published information that attributed other causality to various fish kills that they described. When attempting to make the difficult step from correlation to implication of causality in a field setting, the available evidence for multiple causative factors should be considered-especially when dealing with a topic that has significant implications for policy makers, scientists, and the general public. Accordingly, formal correction of the Paerl et al. paper is necessary because of the authors' (1) lack of depth profiles of dissolved oxygen (DO) data to support any of their conclusions about kills of surface-schooling fish, (2) use of unrecorded or nonexistent fish kill data, as well as misconveyance of fish kills as reported in the State database upon which they relied, (3) apparent lack of understanding about the behavior of resident fish population~, (4) misuse of literature citations, (5) omission of a large body of peer-reviewed, published information on the same fish kills (1995-1996), and (6) lack of any supporting statistical analyses to demonstrate relationships among field dissolved oxygen, nutrient, and fish kill data. Lack of supporting dissolved oxygenlfish kill data and literature. The only DO data that Paerl et al. presented were from the bottom water. A much more complete database (laboratory of J.M.B. & H.B.G.; summary reports covering the period 1993 through 1996, available from the Division of Water Quality [DWQ] of the North Carolina Department of Environment & Natural Resources [NC DENR]) than that discussed by Paerl et al. contains information for physical, chemical, and biological factors on the mesohaline Neuse. This area (Fig. 1) includes the segments addressed in Paerl et al. and the segments where major fish kills historically have occurred in that estuary. Here we compare Paerl et al.'s contentions, based on biweekly data from 6 to 8 mainstem Neuse stations, with the data of our State-certified laboratory, including DO depth profiles. We report weekly data from 6 stations in the mainstem, mesohaline Neuse (Flanners Beach/Kennel Beach to MinnesottKherry Point), as well as data from freshwater segments following a major storm in late summer 1996. These data show that in summers of averageprecipitation years without hurricanes, hypoxic waters in the Neuse Estuary generally were constrained to the bottom third of the water column (Figs. 2 to 4 of this paper; hypoxia considered as in Paerl et al.). Paerl et al. stated that they used a fish kill database from NC DWQ (formerly the Division of Environmental Management [DEM] of the NC Department of Environment, Health & Natural Resources [NC DEHNR]) as