Flow-driven variation in intertidal community structure in a Maine estuary

Ecology, June, 1998 by George H. Leonard, Jonathan M. Levine, Paul R. Schmidt, Mark D. Bertness

INTRODUCTION

Understanding the factors that determine community structure is one of the most important issues in ecology. Although many factors have been emphasized (e.g., physical stress [Davidson and Andrewartha 1948], predation [Paine 1966], competition [Connell 1961], recruitment [Roughgarden et al. 1988], and primary productivity [Slobodkin 1960]) their combined influences remain poorly understood. Models of community structure fall into two broad categories. Nutrient/productivity models (Hairston et al. 1960, Fretwell 1977, Oksanen et al. 1981) and environmental stress models (Menge and Sutherland 1976, 1987) both seek to explain similar patterns but do so from different viewpoints. Nutrient/productivity models predict alternating control by resources or consumers depending on the level of in situ productivity and the number of trophic levels in the food web. In contrast, environmental stress models predict control by competition or consumers depending on the level of stress and whether prey (prey stress models) or predators (consumer stress models) are more strongly affected by this stress.

Nutrient/productivity models, originally developed in terrestrial systems (Hairston et al. 1960), have been applied most successfully in aquatic habitats. Manipulations in streams support the predictions of these simple models (Wootton and Power 1993) and those in lakes suggest that bottom-up and top-down processes are often closely interrelated (Bergquist and Carpenter 1986, Carpenter et al. 1987). Strong predation by fishes in these systems can alter local productivity by changing the relative abundance of lower trophic levels and accelerating nutrient regeneration rates (Leibold 1989, Vanni and Findlay 1990, Vanni and Layne 1997, Vanni et al. 1997). Because these are relatively closed systems, primary productivity is largely intrinsically controlled and results in a tight coupling between bottom-up and top-down forces through biological feedbacks. Consumer stress models, in contrast, have been best studied and applied in marine systems, especially rocky intertidal habitats (Menge 1978a, b). Gradients of physical stress often affect predator mobility and per capita feeding rates and shift the primary factor governing communities from predation to competition (Menge and Sutherland 1987).

The interactions between environmental stress and primary productivity in governing community structure are generally unclear (Menge and Olson 1990, Persson et al. 1996). For example, Wootton et al. (1996) found that flooding disturbance in streams, instead of shortening food chains, actually lengthens them by differentially affecting predator-resistant and predator-susceptible prey. We suggest that predictions of both nutrient/productivity and consumer stress models may apply in systems where the physical process governing variation in resources also act as a stress (sensu Menge and Sutherland 1987) for consumers. Specifically, low productivity habitats whose consumers are also subjected to low physical stress may be structured by predation. In contrast, high productivity habitats whose consumers also experience high physical stress may be governed by strong bottom-up processes. This type of linkage between these models has not been considered previously because the physical processes operating to set nutrients in many systems do not directly affect top predators. For example, in terrestrial systems, nutrient input is largely governed by small-scale effects intimately associated with litterfall and decomposition processes (Wiegert and Owen 1971, Hairston and Hairston 1993). In lakes, primary productivity potential is largely determined by seasonal weather conditions and local runoff (Carpenter and Kitchell 1987, Persson et al. 1992). In rivers, productivity potential is determined by flooding events that set the stage upon which fish predators then act (Power 1990). In none of these cases do these processes seem to influence consumer behavior directly.

Marine systems may provide a striking contrast to these examples and provide insight into the relationship between nutrient/productivity and consumer stress models. Until recently (Menge 1992, Menge et al. 1994, 1996) there has been little work on the interplay between resources and consumers in marine systems. This may be due to the large spatial scale over which variation in primary production operates in the sea as well as the success of consumer stress models in explaining many patterns on open coast intertidal habitats (Menge and Sutherland 1987). Marine systems, however, may be ideal to examine the relative role of bottom-up and top-down effects because of the way physical factors can drive both these processes. Many wave-protected marine systems, such as estuaries and nearshore habitats, are subjected to variable hydrodynamics driven largely by tides and currents. These processes can transport larvae (Genin et al. 1986, Roughgarden et al. 1988, Bertness et al. 1991, Pawlik et al. 1991), suspended food particles (Sebens 1984, Grizzle and Morin 1989, Sanford et al. 1994), and nutrients and gases (Wheeler 1980, Gerard 1982, Koehl and Alberte 1988, Carpenter et al. 1991, Patterson et al. 1991) to benthic habitats. These hydrodynamic conditions can also influence predator foraging efficiency (Kitching et al. 1959, Menge 1978a, b, Burrows and Hughes 1989) and affect top-down control. This suggests that the relative contribution of bottom-up and top-down forces in some marine systems may be related to local hydrodynamic conditions and their role as a "stress" for predators and prey.