Biotic and abiotic controls on the ecosystem significance of consumer excretion in two contrasting tropical streams

1. Excretion of nitrogen (N) and phosphorus (P) is a direct and potentially important role for aquatic consumers in nutrient cycling that has recently garnered increased attention. The ecosystem‐level significance of excreted nutrients depends on a suite of abiotic and biotic factors, however, and few studies have coupled measurements of excretion with consideration of its likely importance for whole‐system nutrient fluxes. 2. We measured rates and ratios of N and P excretion by shrimps (Xiphocaris elongata and Atya spp.) in two tropical streams that differed strongly in shrimp biomass because a waterfall excluded predatory fish from one site. We also made measurements of shrimp and basal resource carbon (C), N and P content and estimated shrimp densities and ecosystem‐level N and P excretion and uptake. Finally, we used a 3‐year record of discharge and NH₄‐N concentration in the high‐biomass stream to estimate temporal variation in the distance required for excretion to turn over the ambient NH₄‐N pool. 3. Per cent C, N, and P body content of Xiphocaris was significantly higher than that of Atya. Only per cent P body content showed significant negative relationships with body mass. C:N of Atya increased significantly with body mass and was higher than that of Xiphocaris. N : P of Xiphocaris was significantly higher than that of Atya. 4. Excretion rates ranged from 0.16–3.80 μmol NH₄‐N shrimp^?1 h^?1, 0.23–5.76 μmol total dissolved nitrogen (TDN) shrimp^?1 h^?1 and 0.002–0.186 μmol total dissolved phosphorus (TDP) shrimp^?1 h^?1. Body size was generally a strong predictor of excretion rates in both taxa, differing between Xiphocaris and Atya for TDP but not NH₄‐N and TDN. Excretion rates showed statistically significant but weak relationships with body content stoichiometry. 5. Large between‐stream differences in shrimp biomass drove differences in total excretion by the two shrimp communities (22.3 versus 0.20 μmol NH₄‐N m^?2 h^?1, 37.5 versus 0.26 μmol TDN m^?2 h^?1 and 1.1 versus 0.015 μmol TDP m^?2 h^?1), equivalent to 21% and 0.5% of NH₄‐N uptake and 5% and <0.1% of P uptake measured in the high‐ and low‐biomass stream, respectively. Distances required for excretion to turn over the ambient NH₄‐N pool varied more than a hundredfold over the 3‐year record in the high‐shrimp stream, driven by variability in discharge and NH₄‐N concentration. 6. Our results underscore the importance of both biotic and abiotic factors in controlling consumer excretion and its significance for nutrient cycling in aquatic ecosystems. Differences in community‐level excretion rates were related to spatial patterns in shrimp biomass dictated by geomorphology and the presence of predators. Abiotic factors also had important effects through temporal patterns in discharge and nutrient concentrations. Future excretion studies that focus on nutrient cycling should consider both biotic and abiotic factors in assessing the significance of consumer excretion in aquatic ecosystems.     

Patterns of litter production across a salinity gradient in a Pterocarpus officinalis tropical wetland

Historically, Pterocarpus officinalisJacq. (Leguminoseae) dominated freshwater wetlands in the coastal plains of Puerto Rico, but deforestation has reduced its distribution to small patches adjacent to mangrove forests in areas of higher salinity. The objective of this study was to determine how a gradient in soil salinity affected litter, flower, and fruit production in a Pterocarpus officinalis.Three 100 m² plots were established in each of three sites along a salinity gradient: pasture/Pterocarpus edge (low salinity, mean salinity at 60 cm–9.7 g Kg^–1), Pterocarpus forest (intermediate salinity, 11.5 g Kg^–1) and a Pterocarpus/mangrove ecotone (high salinity, 15.0 g Kg^–1). Across this gradient, P. officinalis accounted for 100% of the relative basal area in the low and intermediate sites and 43% in the high salinity site which was domimated by Laguncularia racemosa. The basal area of P. officinalis decreased along the gradient from 73.5 m² ha^–1 in the low salinity site to 42.0 m² ha^–1 in the high salinity site. Litterfall was sampled on average every 23 days in 45 0.25 m² traps (5 traps per plot) for two years. Annual litterfall for the forest was 11.9 Mg ha^–1 yr^–1. Peaks in litterfall were associated with high precipitation in May 1995 and tropical storms in September 1995. Leaf fall of P. officinalis was significantly higher in the low salinity site (4.8 Mg ha^–1 yr^–1) than the high salinity site (1.8 Mg ha^–1 yr^–1), but total stand litterfall was greatest in the area of high salinity due to the greater contribution of L. racemosa. Pterocarpus flower and fruit production was approximately 10 times greater in low and intermediate salinity sites in comparison with the high salinity site. An increase in global temperature, will lead to higher sea level and higher soil salinity in costal wetlands. To conserve this wetland forest type it is critical to expand the distribution into areas of lower salinity where this species occurred historically.     

Leaf growth and demography of the rheophytic fern Thelypteris angustifolia (Willdenow) Proctor in a Puerto Rican rainforest

Morphological and temporal aspects of the growth of leaves of Thelypteris angustifolia (Willdenow) Proctor are under study along a 400-m section of the Sonadora River at the El Verde Field Station in the Luquillo Experimental Forest of Puerto Rico. Emergence, expansion and growth of leaves of 149 sporophytes have been observed approximately mid-month every January, May and September since September 1991. Results of the first four years of this long-term study of T. angustifolia are reported here. Although they are land plants, sporophytes of T. angustifolia are rheophytic, subject to intermittent flooding and submerged approximately 7% of the time. The pinnate leaves of T. angustifolia exhibit sterile-fertile leaf dimorphy. Only 7.6% of the leaves in the the sample were fertile. The petiole of a fertile leaf was 41% longer than that of a sterile leaf while leaf blades were the same length. Leaves expanded at a mean rate of 1.7 cm per day, maturing in 29.7 days. Mean leaf life span was 10.8 months with a maximum of 26 months. Sterile leaf longevity (11.0 months) was longer than that of the fertile leaf (9.6 months). Leaves were produced at a mean rate of 4.7 leaves per plant per year. Leaf damage occurred on 38% of the leaves observed. Net leaf counts for individual sporophytes ranged from one to eight with a mean of 3.0 leaves per plant. Fertile plants had higher mean leaf counts (4.1) than plants without fertile leaves (2.8). Leaf production, although possibly triggered by increasing daylength, also paralleled seasonal increases in rainfall during the wetter mid-May/September months. Increased leaf mortality occurred during the drier mid-January/May period.     

Reproduction in an aggregating grouper,the red hind,epinephelus guttatus

Synopsis We examined the reproductive pattern of an aggregating grouper, the red hind,Epinephelus guttatus, in Puerto Rico. Macroscopic and histological examination of gonads confirmed that, although mature, ripe ovaries are found over a three-month period, spawning activity is limited to about 2 weeks each year. Females are determinate spawners and individuals are able to spawn more than once during the course of the annual spawning season. The red hind is protogynous and 50% maturity is attained at 215 mm fork length. In western Puerto Rico, spawning occurs in aggregations at several sites within loosely defined areas located towards the edge of the insular platform. Sex ratios of individuals taken by hook and line at one of the aggregation areas over a consecutive six-year period, suggest considerable intra- and inter-annual variation most likely attributable to a combination of differential ingress and egress by males and females during the course of an aggregation and to fluctuations in recruitment, differential mortality by sex and sex change among years. A comparative assessment of the reproductive patterns of seven western AtlanticEpinephelus spp. suggests that aggregation-spawning is associated with medium- to large-sized groupers, while smaller groupers do not aggregate. Mating systems vary among congeners in association with trends in male-female size dimorphism, sexual pattern and sperm competition. The short-term, localized, nature of spawning aggregations renders species with this reproductive mode particularly vulnerable to fishing pressure at spawning sites. Aggregating species, therefore, may require special management consideration.     

A new genus and species of 'giant hutia' (Tainotherium valei) from the Quaternary of Puerto Rico: an extinct arboreal quadruped?

A large incomplete rodent femur from a Quaternary cave deposit near Barahona, Puerto Rico, is established as the holotype of Tainotherium valei , a new extinct genus and species. Although biogeographic and body size similarities suggest that it may be related to the Puerto Rican giant hutia Elasmodontomys , the Antillean large-bodied rodent family Heptaxodontidae is now interpreted as invalid, and it is impossible to assign Tainotherium to a particular caviomorph family in the absence of associated craniodental material. Tainotherium differs from other West Indian species in possessing a large femoral head, a proximally angled femoral neck, a short greater trochanter and a medially positioned lesser trochanter unconnected by an intertrochanteric crest, and a transversely flattened, anteroposteriorly bowed shaft lacking well-defined ridges. These characters are all associated with arboreal life habits in other mammal groups. The Puerto Rican land mammal fauna was dominated by a rodent radiation occupying a wide variety of niches before human arrival in the West Indies, but although arboreality is correlated with increased likelihood of survival in Quaternary mammalian extinction events, all of this fauna is now extinct. It is unlikely that decreasing aridity and the reduction of Puerto Rican savanna-type environments at the end of the Pleistocene contributed to the extinction of the arboreal Tainotherium , and habitat destruction by pre-Columbian Amerindians may instead have been responsible.     

Modeling the effects of Hurricane Hugo on spatial and temporal variation in primary productivity and soil carbon and nitrogen in the Luquillo Experimental Forest, Puerto Rico

Hurricanes account for much of the spatial and temporal variation in forest productivity and soil organic matter pools in many forest ecosystems. In this study, we used an ecosystem level model, TOPOECO, to simulate the effects of Hurricane Hugo (18 September 1989) on spatial and temporal patterns of gross primary productivity (GPP), net primary productivity (NPP), soil organic carbon (SOC) and nitrogen over the entire Luquillo Experimental Forest (LEF), Puerto Rico, a tropical rainforest. Our simulation results indicated that simulated annual GPP increased by an average of 30% five years after Hugo in the Tabonuco forest at low elevations where there was a fast recovery of the canopy, whereas simulated GPP decreased by an average of 20% in the Palm and Dwarf forests at high elevations as a result of the slow recovery of the canopy. Simulated annual NPP in the Palm and Dwarf forests also did not recover to pre-Hugo levels within 5 years. Simulated storages of SOC, CO₂ emission from decomposition of SOC and total soil nitrogen increased slightly but N mineralization rate increased significantly in all four vegetation types due to the massive input of plant materials from Hugo at low elevations and the slow decomposition at high elevations.