What is a bloom ? A commentary Theodore

There are at least eight different modes and mechanisms by which harmful phytoplankton species can cause mortality, physiological impairment, or other negative in situ effects. Their distinction from nonharmful phytoplankton taxa is clearly warranted. Increasing use of bloom descriptors such as “exceptional,” “unusual,” “nuisance,” and subjective reference to specific occurrences as blooms reveal widespread confusion concerning: what is a bloom?, how is it to be defined?, what is harmful?, and what distinguishes a harmful bloom from other blooms? Such efforts have been influenced by comparison with perceived spring diatom bloom characteristics. Examples, selected from among harmful bloom events, arc presented to support the view that the classical focus on the spring diatom bloom has significant conceptual and operational biases as to what coqstitutcs a bloom. This and the inexact criteria used to define a bloom compromise research on both harmful algal blooms and phytoplankton blooms in general. The subjective, differing, and arbitrary criteria used to define blooms and their presumed ecological consequences need to be replaced by a quantitatively based, ecological classification of the various types of phytoplankton blooms. The issue of what constitutes a bloom is more than simply a biomass issue. The term “harmful” algal blooms (HABs), referred to as “red tides” in the older literature, has been applied to a class of blooms increasingly thought to have unique properties (Smayda 1997). This emergent bloom phenomenon, rapidly expanding in global coastal waters (Anderson 1989; Smayda 1990; Hallegraeff 1993), has many puzzling aspects. Efforts to understand these and to develop mitigation strategies require clarification of what is a harm&! bloom? Evaluation of this and the related issues of what is a bloom? and what is meant by “harmful” is needed to circumvent problems associated with application of traditional views as to what constitutes a bloom, e.g. it is a significant population increase, during which the bloom and subordinate species within the community have equivalent ecological and physiological valence, and such blooms intrinsically are beneficial to food-web processes. These characterizations, current paradigms of bloom regulation and dynamics, and conceptual, experimental, and modeling approaches to blooms are based primarily on a diatom template. This reflects the historical focus of marine phytoplankton ecologists on the annual, high biomass, diatom-dominated spring (upwelling) bloom, whereas harmful algal blooms are primarily flagellate events (Sournia 1995). “Miniblooms” of diatoms developing during other seasons in response to episodic, “new” nutrient pulses are usually ignored. Likewise, blooms of other phylogenetic groups traditionally have been dismissed as ephemeral, transient pulses or, as in the case of red tides, as anomalous rogue blooms of peripheral interest. Rarely are the frequencies and trophodynamic impacts of such blooms considered. The dogma which seems to have emerged is that the spring diatom bloom is the only one worthy of serious study, that it is the major bloom event driving marine trophodynamics, and I.hat resolution of its dynamics will generally explain nutrient-regulated fertility and bloom dynamics. This restricted focus and the spring diatom bloom bias have a practical explanation. Methodological constraints make it difficult to deal with the successions of bloom and subordinate species occurring during bloom events, and their variable ecophysiology, bloom magnitude potential, and suitability as prey. Inherently, particularly during blooms, these interrelated successional features have both causes and consequences. Quantitative measurements of bloom processes as a function of these parameters are not only methodologically difficult, but usually further constrained by the need for high popuhtion abundance. These analytical problems are classically accommodated by applying a reductionist, whole-community approach, with community size-fractionation a partial refinement (Malone 1980). Bloom abundance and fluctuations are then commonly expressed in terms of community biomass, with chlorophyll amount usually the index of abundance against which total-community rate processes, such as productivity, are often normalized. Of the bloom events in the sea, the high abundance and monotonous diatom flora of the annual spring (upwelling) bloom historically have best satisfied these analytical needs. This largely explains our biased understanding of blooms, based primarily on prcaximate, whole-community approaches, with neglect of the organismal approaches essential to elucidation of bloom events. Some phytoplankton ecologists, who typically focus on rnass balance dynamics and primary production, have slowly come to recognize what organismally focused ecologists accept as a truism--the trophic consequences of blooms should vary with the bloom species. This is particularly evident to those who investigate HAB