The 326 ha Río Icacos watershed in the tropical wet forest of the Luquillo Mountains, northeastern Puerto Rico, is underlain by granodiorite bedrock with weathering rates among the highest in the world. We pooled stream chemistry and total suspended sediment (TSS) data sets from three discrete periods: 1983–1987, 1991–1997, and 2000–2008. During this period three major hurricanes crossed the site: Hugo in 1989, Hortense in 1996, and Georges in 1998. Stream chemistry reflects sea salt inputs (Na, Cl, and SO4), and high weathering rates of the granodiorite (Ca, Mg, Si, and alkalinity). During rainfall, stream composition shifts toward that of precipitation, diluting 90% or more in the largest storms, but maintains a biogeochemical watershed signal marked by elevated K and dissolved organic carbon (DOC) concentration. DOC exhibits an unusual “boomerang” pattern, initially increasing with flow but then decreasing at the highest flows as it becomes depleted and/or vigorous overland flow minimizes contact with watershed surfaces. TSS increased markedly with discharge (power function slope 1.54), reflecting the erosive power of large storms in a landslide‐prone landscape. The relations of TSS and most solute concentrations with stream discharge were stable through time, suggesting minimal long‐term effects from repeated hurricane disturbance. Nitrate concentration, however, increased about threefold in response to hurricanes then returned to baseline over several years following a pseudo first‐order decay pattern. The combined data sets provide insight about important hydrologic pathways, a long‐term perspective to assess response to hurricanes, and a framework to evaluate future climate change in tropical ecosystems.

Anthropogenic actions are altering fluxes of nitrogen (N) in the biosphere at unprecedented rates. Efforts to study these impacts have concentrated in the Northern hemisphere, where experimental data are available. In tropical developing countries, however, experimental studies are lacking. This paper summarizes available data and assesses the impacts of human activities on N fluxes in Puerto Rico, a densely populated Caribbean island that has experienced drastic landscape transformations over the last century associated with rapid socioeconomic changes. N yield calculations conducted in several watersheds of different anthropogenic influences revealed that disturbed watersheds export more N per unit area than undisturbed forested watersheds. Export of N from urban watersheds ranged from 4.8 kg ha)1 year)1 in the Rı´o Bayamo´ n watershed to 32.9 kg ha)1 year)1 in the highly urbanized Rı´o Piedras watershed and 33.3 kg ha)1 year)1 in the rural-agricultural Rı´o Grande de An˜ asco watershed. Along with land use, mean annual runoff explained most of the variance in fluvial N yield. Wastewater generated in the San Juan Metropolitan Area receives primary treatment before it is discharged into the Atlantic Ocean. These discharges are N-rich and export large amounts of N to the ocean at a rate of about 140 kg ha)1 year)1. Data on wet deposition of inorganic N (NHþ4 þ NO
3 ) suggest that rates of atmospheric N deposition are increasing in the pristine forests of Puerto Rico. Stationary and mobile sources of NOx (NO+NO2) and N2O generated in the large urban centers may be responsible for this trend. Comprehensive measurements are required in Puerto Rico to quantitatively characterize the local N cycle. More research is required to assess rates of atmospheric N deposition, N fixation in natural and human-dominated landscapes, N-balance associated with food and feed trade, and denitrification.

Here we test the hypothesis that differences in macrobiotic assemblages can lead to differences in the quantity and quality of organic matter in benthic depositional environments among streams in montane Puerto Rico. We experimentally manipulated biota over a 30–40 d period in two streams with distinctly different macrobiotic assemblages: one characterized by high densities of omnivorous shrimps (Decapoda: Atyidae and Xiphocarididae) and no predaceous fishes, and one characterized by low densities of shrimps and the presence of predaceous fishes. To incorporate the natural hydrologic regime and to avoid confounding artifacts associated with cage enclosures/exclosures (e.g., high sedimentation), we used electricity as a mechanism for experimental exclusion, in situ. In each stream, shrimps and/or fishes were excluded from specific areas of rock substrata in four pools using electric “fences” attached to solar-powered fence chargers. In the stream lacking predaceous fishes (Sonadora), the unelectrified control treatment was almost exclusively dominated by high densities of omnivorous shrimps that constantly ingested fine particulate material from rock surfaces. Consequently, the control had significantly lower levels of inorganic sediments, organic material, carbon, and nitrogen than the exclusion treatment, as well as less variability in these parameters. Tenfold more organic material (as ash-free dry mass, AFDM) and fivefold more nitrogen accrued in shrimp exclosures (10.6 g AFDM/m2, 0.2 g N/m2) than in controls (1.1 g AFDM/m2, 0.04 g N/m2). By reducing the quantity of fine particulate organic material and associated nitrogen in benthic environments, omnivorous shrimps potentially affect the supply of this important resource to other trophic levels. The small amount of fine particulate organic matter (FPOM) that remained in control treatments (composed of sparse algal cells) was of higher quality than that in shrimp exclosures. This is evidenced by the significantly lower carbon-to-nitrogen (C/N) ratio (an indicator of food quality, with relatively low C/N indicating higher food quality) in the control relative to the shrimp exclosure treatment. In contrast, the stream with predaceous fishes (Bisley) was characterized by very low numbers of shrimps, and macrobiota had no significant effect on benthic sediments, organic matter, C, N, and C/N. All parameters were highly variable through time, with levels and ranges in variability similar to the shrimp exclusion treatment in the Sonadora. Our experimental results are consistent with findings of an independent survey of six streams in four different drainages. Four streams that had an abundance of omnivorous shrimps, but lacked predaceous fishes, had extremely low levels of fine benthic organic and inorganic material. In contrast, two streams that had low densities of shrimps and contained predaceous fishes had significantly higher levels. Results show a strong linkage between species and ecosystem characteristics: interstream differences in the quantity and quality of fine benthic organic matter resources were determined by the nature of the macrobiotic assemblage. Furthermore, patterns in the distribution of shrimp assemblages reflected landscape patterns in the benthic depositional environment among streams.

Humid tropical forests are often characterized by large nitrogen (N) pools, and are known to have large potential N losses. Although rarely measured, tropical forests likely maintain considerable biological N fixation (BNF) to balance N losses. We estimated inputs of N via BNF by free-living microbes for two tropical forests in Puerto Rico, and assessed the response to increased N availability using an on-going N fertilization experiment. Nitrogenase activity was measured across forest strata, including the soil, forest floor, mosses, canopy epiphylls, and lichens using acetylene (C2H2) reduction assays. BNF varied significantly among ecosystem compartments in both forests. Mosses had the highest rates of nitrogenase activity per gram of sample, with 11 ± 6 nmol C2H2 reduced/g dry weight/h (mean ± SE) in a lower elevation forest, and 6 ± 1 nmol C2H2/g/h in an upper elevation forest. We calculated potential N fluxes via BNF to each forest compartment using surveys of standing stocks. Soils and mosses provided the largest potential inputs of N via BNF to these ecosystems. Summing all components, total background BNF inputs were 120 ± 29 lg N/m2/h in the lower elevation forest, and 95 ± 15 lg N/m2/h in the upper elevation forest, with added N significantly suppressing BNF in soils and forest floor. Moisture content was significantly positively correlated with BNF rates for soils and the forest floor. We conclude that BNF is an active biological process across forest strata for these tropical forests, and is likely to be sensitive to increases in N deposition in tropical regions.