Alonso Ramírez

A TEST OF TOP‐DOWN AND BOTTOM‐UP CONTROL IN A DETRITUS‐BASED FOOD WEB

In food webs based on primary production, biomass of organisms within trophic levels can be simultaneously controlled by resources (bottom-up) and consumers (top-down). In contrast, very little is known about top-down and bottom-up control in detritus-based food webs. Here, we tested whether exclusion of macroconsumers (fishes and shrimps) and/or phosphorus (P) addition affected insect detritivore biomass and decay rate and quality of leaf detritus in a tropical stream. Four treatments were established in a third-order stream in Costa Rica: (1) macroconsumers present, ambient P; (2) macrocon- sumers excluded, ambient P; (3) macroconsumers present, P added; (4) macroconsumers excluded, P added. Biomass of insect larvae inhabiting leaf packs and mass loss of leaves were measured after 7 and 10 d in situ. After 10 d, biomass and density of insect larvae increased as a result of both P addition and exclusion of macroconsumers. Chironomids (Diptera, Chironominae) were the dominant detritivores in leaf packs, comprising 51-80% of total invertebrate biomass and were responsible for the observed treatment effects. Mass loss rates of leaf packs were accelerated by the presence of macroconsumers but not by P addition. Positive effects of P on insects presumably occurred through increased microbial carbon relative to leaf carbon. However, percentage nitrogen (N), C:N, and fungal biomass of leaves were not affected by either experimental treatment. Laboratory growth studies corroborated positive effects of P on chironomids: growth rates were higher in high-P treatments (high-P stream water and low-P stream water with P added) vs. low-P stream water. However, no differences in the in situ growth rates were observed between a high- and a low-P stream. The relative importance of top-down and bottom-up effects was evaluated using several indices. Loss rates of organic matter were affected more by top-down effects of macro- consumers than by bottom-up effects of P. Macroconsumers had negative effects on two trophic levels, contrary to food-web theory predicting alternating negative and positive effects. Positive bottom-up effects of P on insect biomass were greater than negative top- down effects of macroconsumers. In addition, P effects on invertebrates were similar in direction but greater in magnitude than previously published effects of nutrients on con- sumers in food webs based on primary producers. These results suggest that the impacts of nutrient enrichment on detritivores may be as great or greater than those previously observed on herbivores.

Invertebrate drift and benthic community dynamics in a lowland neotropical stream, Costa Rica

In this study we quantified invertebrate drift and related it to the structure of the benthic community, over a 6–8 month period, in a 4th-order tropical stream in Costa Rica. Relative to reports from similar-sized temperate and tropical streams, drift densities were high (2-fold greater: mean 11.2 m−3; range 2.5–25 m−3), and benthic insect densities were relatively low (>3-fold lower: mean 890 m−2; range 228–1504 m−2). Drift was dominated by larval shrimps that represented more than 70% of total drift on any given date; the remaining 30% was composed of 54 insect taxa. Among insects, Simuliidae and Chironomidae (Diptera) and Baetidae, Leptohyphes and Tricorythodes (Ephemeroptera) comprised 24% of total drift. Drift periodicity was strongly nocturnal, with peaks at 18:00 h (sunset) and 03:00 h. Our results, and those of previous experiments in the study stream, suggest that nighttime drift is driven by the presence of predatory diurnal drift-feeding fishes and nocturnal adult shrimps. There were no clear seasonal patterns over both ‘dry’ and wet seasons, suggesting that benthic communities are subject to similar stresses throughout the year, and that populations grow and reproduce continuously.

Do macroconsumers affect insect responses to a natural stream phosphorus gradient

We assessed the role of macroconsumers (e.g. fishes and shrimps) in affecting the response of insect assemblages to a natural phosphorus gradient formed by six streams with different phosphorus levels (range 12–350 μg/l). We hypothesized that insect responses to the phosphorus gradient would be strongest in the absence of macroconsumers. Within each stream, macroconsumers were allowed access to, or were excluded from, leaf packs using electric `fences. Macroconsumers did not have significant effects on insects, but there was a significant phosphorus effect. Insect assemblages in high-phosphorus streams had 3–12-fold greater biomass and 3–11-fold greater abundance than assemblages in low-phosphorus streams. We also found that insect responses to phosphorus were more variable when assessed on the natural benthic substrate (e.g. mixed detritus) than in standardized leaf packs. In both substrates, the relationship between insects and phosphorus was not linear: abundance and biomass increased with phosphorus concentration to an asymptote. This suggests that insects were responding to a food resource gradient (e.g. fungi and bacteria). The Michaelis–Menten model provided a good fit for the relationship between insects and phosphorus concentrations, with half-saturation constants ranging from 12 to 60 μg SRP/l. The asymptotic relationship observed between phosphorus and insects suggests that phosphorus saturation occurred above a threshold of ca. 100 μg SRP/l. Our results provide support for the hypothesis that detritus-based food webs are mainly controlled by bottom-up forces.