Feces in Aquatic Ecosystems

Think of trophic levels, and what probably comes to mind is an illustration from a biology text showing a pyramid, with solar energy trapped by photosynthesizing plants on the bottom, plants fed upon by herbivores in the middle, and herbivores eaten by carnivores on top. These models may also show arrows indicating that feces and dead plants and animal bodies provide inputs to the detritus pool, illustrating how this organic matter is recycled by microorganisms, and how at all trophic levels, aerobic respiration results in the loss of energy from the ecosystem. Such conceptual models provide a basis for energy budgets to be investigated, but the focus of early studies was on energy production and consumption by plants and animals. In many contemporary studies scientists continue to concentrate on feeding, investigating food webs, optimal foraging, predator–prey interaction, or the dynamics of functional feeding groups. Partly this is because the diet of animals, and the mode of food capture, can be used to model complex processes, and the strategies and tactics differ among populations. Much less attention has been paid to the role of feces in ecosystems, yet fecal pellets are often very abundant, represent a repackaging of available organic matter, and are readily transported. Aquatic plants photosynthesize only where light penetrates, and the photic zone makes up only a tiny fraction of the depth of oceans, but a greater fraction of the depth in most lakes. Primary production in the surface waters supports the biotic community of the photic zone, and feces and dead matter produced there descend through the water column. Feces thus provide an important flux of carbon from surface to deeper waters in oceans, and a similar vertical flux occurs in lakes. Unlike oceans and lakes, rivers receive much organic matter from terrestrial sources, and animals capture this from suspension or after it becomes deposited. Feces are carried horizontally by the current in rivers, and the significance of this transport has only recently been recognized. In this article, we discuss the fate of fecal pellets in aquatic ecosystems, particularly with respect to vertical and horizontal flux. First we need to know more about the feces of aquatic animals. Almost all aquatic animals have guts into which gathered, or captured, food is ingested. Within the gut the food is subjected to abrasion, changes in pH, and the action of enzymes, all of which contribute to the breakdown of organic matter into labile compounds that are taken up across the walls of cells lining the gut. The gut lumen is thus best regarded as a space where material is concentrated and converted to molecules that can pass across the gut wall and thus be used in metabolism. Viewed in this way, ingested food is “external” to the animal, as only breakdown products enter animal tissues, and materials excreted into the gut lumen are also passed out from the animal with the egested feces. The egesta usually look very different from the materials ingested, although the extent of the difference varies with the efficiency with which digestion occurs. Feces pass directly to the substratum or into the water column, and some pass from the water column to the substratum after sedimentation. Close examination of the feces of most aquatic animals shows them to contain fragments of the undigested diet. These include the remains of plant cells, pieces of exoskeleton from invertebrates, and much unidentifiable organic matter, the extent of each varying from animal to animal. Ingested living organisms are also egested. For example, some invertebrates survive passage through the gut of bottom-feeding fish (Aarnio and Bonsdorff 1997), and the egestion of live snails by fish may facilitate their spread (Haynes et al. 1985). The feces of planktonic crustaceans, and other animals feeding on single-celled algae, often contain many living algal cells (Wotton 1994). These cells pass through the gut without being affected too adversely, while others are digested. Although bacteria are lysed in large numbers (Fenchel 1970, Wotton 1994), many millions of microorganisms also pass through the gut of animals, apparently little harmed by digestion (Wotton 1994). The degree of resistance of bacteria to lysis varies markedly from strain to strain, with Gram-positive bacteria always resistant (Plante 2000). Some bacteria thrive within the gut of aquatic animals (e.g., in crustaceans, mollusks, and echinoderms; Harris 1993) and thus contribute exudates, living cells, and cell fragments to the egesta. Feces of aquatic animals are typically bound into discrete pellets (Figure 1a), although some feces are diffuse. Fecal pellets usually have a shape characteristic of each animal, and they may be spherical, cylindrical, elongated, or formed into strings (Ladle and Griffiths 1980, Ladle et al. 1987, Noji 1991, Ward et al. 1994). The size of pellets varies both among and within taxa, depending largely on animal size. Protist “pellets” are as small as 6 µm × 9 µm (Stoecker 1984), whereas polychaete worms, tunicates, and euphausiid shrimp (krill) produce pellets several millimeters long (Taghon et al. 1984, Deibel 1990, Gonzalez 1992); fish strings can be longer still. It would be expected that the largest feces are produced by whales, but Best et al. (1995) describe the feces of humpback whales as “a stream of particulate matter” often in an oily matrix. Planktonic crustaceans have a peritrophic membrane, which is a thin tubular sheet secreted in the midgut as a means of preventing abrasion of the gut wall. This acts as a distinct wrapper for feces (Lautenschlager et al. 1978, Lampitt et al. 1990, Gonzalez 1992). The feces of vertebrates and most invertebrates do not have such an external covering and are bound together with mucus, often remaining as discrete entities for days or weeks. Many animals have mucus-secreting cells lining the gut (derived from cells used in distant ancestry to produce mucus for protection or locomotion). Mucus is used to protect the gut wall from damage by abrasive food items, and also as a lubricant (Allen 1981). However, aquatic insect larvae often have a slow-moving peritrophic membrane surrounding the rapidly moving gut contents, so it is surprising that the fecal pellets of many larvae are bound with mucus (Shepard and Minshall 1981, Wotton 1994). This must be either acquired from the environment or secreted by captured organisms on passage through the gut. Many marine animals (e.g., comb jellies, bivalve mollusks, and salps) use mucus as a means of trapping food or as a means of lubricating food passage (e.g., gastropods), and this exudate is then ingested together with the food. Mucus secreted over the body surface and used for protection, or lubrication in locomotion, is also likely to be ingested from time to time. Stains such as alcian blue readily reveal the presence of mucus in pellets, but scanning electron micrographs often show very small amounts of this binding material, which becomes dehydrated during specimen preparation (Figure 1b). The number and quality of fecal pellets produced by aquatic animals varies with the feeding strategy employed. Predators convert animals into smaller particles of detritus (Figure 2), and as they have high assimilation efficiencies, only a small number of feces are produced by members of this functional feeding group. Although predators thus have an important top-down effect on community structure, the contribution of their fecal material to organic matter dynamics is small unless they are locally abundant. Shredders feed largely on living or dead plants and, like predators, reduce overall particle size (Figure 2). As the assimilation efficiency of herbivores is generally lower than that of carnivores, the number of pellets produced by shredders is higher than that of predators. Scrapers feed on biofilms that are gathered into the gut. Often large numbers of fecal pellets are produced, and particle size is reduced if biofilm, rather than its constituents, is considered (Figure 2). Scrapers such as gastropod mollusks are found in very high densities on the surface of mud and other substrata, and the presence of more than 100 pellets cm−2 (i.e., more than one million per square meter) is common (Ward et al. 1994). Collectors, and especially suspension feeders, gather many small particles, colloids, and dissolved organic matter and transform the material into much larger fecal aggregates (Figure 2). Bivalve mollusks are suspension feeders that can dominate the substratum in many types of water body, and they trap particles over the surface of their gills. Once collected, particles are coated in mucus and transferred to the labial palps, where some are selected for ingestion and others rejected. Rejected particles, and those that are drawn into the mantle cavity in excessive quantities, are bound with mucus and expelled from within the shell valves as pseudofeces (Ward and MacDonald 1996). Feces and pseudofeces thus look similar and are found mixed together in the vicinity of bivalves (Strayer et al. 1999). Pseudofeces are a true case of repackaging, as they are not affected by digestion. Typically, suspension feeders produce large numbers of fecal pellets, for example, 3.2 pellets h−1 for copepods (Griffin 2000), and extrapolated means of 575 and 737 pellets d−1 for black fly larvae (Wotton et al. 1998, Malmqvist et al. 2001). As black fly larvae are often found in densities exceeding 100,000 m−2, the total number of pellets they produce is enormous (Wotton et al. 1998). Abundant macrozooplankton also produce fecal pellets in large numbers: Alldredge et al. (1987) found more than 9 × 104 pellets m−3 in the surface waters off southern California. It is the abundance and feeding rate of suspension feeders that make them such significant transformers of organic matter. Although fecal pellets of many kinds abound and are utilized by communities over the substratum, the pellets of suspension feeders are not only extremely numerous but also provide the translocation of repackaged organic matter (Le Fèvre et al. 1998) from one location to another as they sink or are carried by currents. Two mechanisms are especially significant: the vertical flux of fecal pellets in oceans and lakes, and the horizontal transport of fecal pellets in rivers. Feces are egested in the water column by planktonic animals or swept up from the substratum after egestion by benthic animals. Whatever their origins, the sinking rate of fecal pellets in water is often rapid (Figure 3) and depends on pellet size, shape, and surface area (Smayda 1969, Fowler and Small 1972, Ladle et al. 1987, Viitasalo et al. 1999). Sinking rate also depends on the pellets' specific gravity, varying with the food ingested, presence of mineral particles, degree of compression, and rate of decomposition (Smayda 1969, Butler and Dam 1994, Wotton 1994, Hansen et al. 1996, Yoon et al. 1996). Where the water is shallow, there is a large input of fecal pellets to the substratum, but a proportion of the pellets produced by planktonic animals is lost from the photic zone in deep waters. This is particularly true of the larger pellets of animals such as euphausiids (Keck and Wassmann 1996, Gonzalez et al. 2000). “Pellets” produced by protists are very small and are likely to sink slowly, but they are often associated with detrital aggregates that settle rapidly during blooms in spring and summer. All small pellets (like those produced by zooplankton greater than 200 µm in length) are likely to remain within stratified photic zones (Small et al. 1987, Viitasalo et al. 1999), unless they become associated with aggregates. For example, only 5% of copepod feces produced in the photic zone of the Humboldt Current were found in sediment traps at 300 m (Gonzalez et al. 2000), and Gonzalez (1992) found a more than 99% decrease in fecal strings of euphausiids from the upper 50 m to depths of m in waters. flux of fecal pellets in the of up to d−1 in the 50 m et al. are usually very much less than with only a tiny fraction to organic carbon at depths of m are about in oceans, and in oceans and the area of oceans, the flux of organic carbon to deep water is and has been at and much of this material of fecal pellets, both and in aggregates of is not In stratified oceans, the loss of larger pellets and aggregates from the photic zone means of organic matter, and are only by However, in stratified lakes, the by of pellets at depth are to surface waters at of Some mechanisms are by et al. 1999), and this is likely pellets become diffuse. The of pellets them to up into of smaller and sinking but breakdown to be very rapid for this to that zooplankton fecal pellets are by although the rate of breakdown is affected by the of the diet et al. 1996). However, bacteria associated with are more than those living in the water and this is to those bacteria bound within fecal Some copepods up pellets while little of their et al. 1990, Noji et al. and the fragments thus sink at a rate than feces. the copepods the of pellets, which thus become more and that sinking then become of greater The pellets of egested into water are by a peritrophic membrane for up to after which time the membrane becomes probably by of the pellets (Lautenschlager et al. and they then become diffuse. of invertebrates, like those of are of and of mucus, microorganisms, and algal As aggregates are it is likely that they sink more than aggregates of the but of detritus are common of the especially at some of the The sinking of fecal pellets is by the to which they have become mechanisms that the sinking rate of fecal pellets are the of that or the presence of et al. 1987, Lampitt et al. 1990, Wassmann 1994, Yoon et al. 1996, Viitasalo et al. 1999). Some crustaceans high in the water column they to surface waters at et al. with the of to pellets from sinking across the the water body is In other feeding in the surface waters the passage of fecal pellets by feeding the surface at and fecal pellets during the after the animals have several in the water column et al. Some fish also to the substratum of at and there and 1994), and fish feces from used in fish over the substratum 1998). living on the on or in rivers fecal pellets that are horizontally by In the pellets are carried by and by the and of and they are with other of the and in large amounts of detritus from feeding and egestion by and suspension This is also true for the that the of and In the feeding of found in very high results in up to of the water being through the animals each In bivalves may by their and et al. a of to for of of the water column per that can be by bivalves in a of different aquatic both marine and a fraction of the particles captured is ingested and with the passed out as but the of both are the of of detritus bound with All fecal pellets on the surface of the substratum are likely to become from time to time and then with other particles in the et al. transport of fecal pellets in rivers is an important of In a 100 m Malmqvist et al. transport of fecal pellets to be and d−1 in each of the high densities of black fly larvae the substratum of and their fecal pellets typically make up to of total matter transport et al. 2001). This with the of particulate organic carbon up of fecal pellets in the surface waters of oceans et al. 2000), with the of to for this proportion found by et al. in and with to for a et al. 1998). Many pellets in the were in and with the being carried to the The significance of horizontal transport and the of fecal pellets produced by black fly larvae was in a of a small stream m from a to the black fly fecal pellets the stream from the but they were produced in such in an m from the that the numbers in transport from pellets (Wotton et al. 1998). were in and this over the larvae were (Wotton et al. 1998). this of in and rivers, many very small particles and dissolved matter would be lost to the The of carbon from the water column to the substratum was likely to be only less than the input of to the the energy to be the one in (Wotton et al. 1998). similar of carbon from suspension to the substratum is likely where horizontal of water pass over of in rivers or the The horizontal transport of fecal pellets in rivers and the vertical transport of fecal pellets in oceans and both provide of repackaged organic matter. translocation comes in small water bodies into which terrestrial animals Feces become available for microorganisms and animals both during transport and after to the Once the pellets of suspension feeders the pellets produced more locally by feeders, and predators. The abundance of pellets depends on the densities of animals within each functional feeding group. the pellets of feeders and suspension feeders as animals feed for long time and so little of their food. to the of organic matter by It is that feces also contain living organisms that pass through the gut little affected by digestion. Many bacteria and algal cells survive gut but so do invertebrates fed upon by fish (Aarnio and Bonsdorff The organisms in egested feces are thus likely to have the for rapid as they are in to of fecal pellets thus occurs from within as as from of the of fecal pellets dissolved organic matter into the water et al. 1999). carbon and are lost from pellets in the days after and if the pellets of many animals remain the size as at egestion but have reduced (Gonzalez and Although a small proportion of pellets is lost from surface it is important to that all is likely to be within the photic zone 1999). are rapidly by microorganisms et al. that may be by the but then and which to a decrease in carbon as carbon is converted to carbon by Feces by or microorganisms have higher than other at in taken from a stream (Ward and and this is typically a that shows a days after egestion and organic materials from the water et al. also become to the binding mucus, and aggregates are In the pellets are rate depending on many although labile may be in days as food as some carbon is by the of microorganisms, being from the water to The pseudofeces of bivalve mollusks to be rapidly and 1997), and this may their of with bivalve feces. The feces egested by some also rapidly (Shepard and Minshall 1981). many of the mucus that pellets are and water the Many aquatic animals feed on fecal pellets and fed several types of benthic animals from the stream we with fecal pellets (Wotton et al. 1998). pellets were produced by black fly larvae feeding on particles in and mixed within the egesta. found of the in many of the the guts of and black fly larvae the in and the guts of larvae and water also many particles (Wotton et al. 1998). fly larvae have been to capture and fecal pellets produced by larvae and this has been regarded as a means of the of feeding (Wotton In some the production of animals depends on of fecal For example, is likely to be a significant in as it the best for the of production that have been and The of fecal pellets and other detritus to the supports the that live with the only other inputs of organic matter being by (Wotton 1994). is also a common of waters and and many animals fish and fecal pellets that are or produced in and that by animals is a more important than breakdown in pellet organic matter over need to more as a significant in aquatic although this to a for the are studies on its being a significant of feeding In most aquatic there is a of In and oceans and lakes, during spring through planktonic most thus production and or the numbers of animals and the of fecal material that they produce et al. is thus an in the particle flux a of fecal pellets and other organic matter at of high production in the photic zone (Le Fèvre et al. 1998). and plants also contribute but plants often in a of organic matter that is to provide energy through the and for the and rivers also receive large inputs from the water body that may be from the or from in this is a in and is over time as and fragments settle to the Fecal pellets from suspension feeders sediment feeding is and the production of suspension feeders is often the of high production in spring and summer. and Minshall that the benthic community feces of many stream invertebrates as food and inputs of terrestrial matter have been that feces a of in and probably provide a and of particulate organic matter in all water In to their as a means of and as food for animals, feces also have other animals that in and similar are likely to fecal for example, pellets produced by larvae are into the in which larvae live and Wotton 1998). for feces is as an It has been found that the feces of contain from one of fish they and et al. 1996). of the can the from feces and are of the presence of the predators et al. 1996). Such do not with all but this is a role of feces that may be in a of predator–prey and be of feeding by animals have on the of aquatic Although the of fecal pellets in vertical transport of organic matter in oceans is the role of feces has been little considered in other aquatic are usually very different in particle size with ingested material and are readily within the water column. are also of and a in In many fecal pellets are very abundant, and we are only to their would like to and for their on an of this which in many of fecal pellets produced by black fly larvae to show their characteristic shape and 50 of of a fecal pellet egested from a black fly The pellet was and and as those in but dehydrated during preparation for scanning electron electron of the of of the of food to feces in animal functional feeding groups. Predators transform animals into smaller fecal pellets, but in as assimilation is feeding on live or dead plant material, produce larger numbers of fecal pellets than predators, but the pellets are smaller in size than the Scrapers the organic covering on and make fecal pellets that are larger than food items, but the are within a of feeders and transform very large numbers of small particles and dissolved matter into much larger fecal The number of pellets produced is often as many suspension feeders and feeders large of food Sinking of fecal pellets as for the Fecal pellets show a of size from to are from Alldredge et al. 1987, 1969, Deibel 1990, Ladle et al. 1987, Fowler and Small 1972, Ladle et al. 1987, Yoon et al. 1996, and et al. of as