Nutrient enrichment in stream ecosystems has the potential to alter food quality (as measured by nutrient : carbon ratios) throughout stream food webs. Ecological stoichiometry predicts that the elemental ratios of consumer species are relatively fixed (homeostatic), while basal resource stoichiometry may vary with nutrient availability. We tested this hypothesis by analyzing carbon (C), nitrogen (N), and phosphorus (P) content of invertebrate consumers and basal resources from a series of streams in lowland Costa Rica that range widely in P-availability (2-135 µg/L soluble reactive phosphorus) due to natural inputs of solute-rich groundwater. P-content of detritus and epilithon increased with stream P levels, exceeding some of the highest values reported in the literature. Invertebrate consumers from multiple taxa and across functional feeding groups increased in P-content in the P-rich stream. While primary consumers typically face the greatest imbalance between body composition and food resources, our results indicate that this imbalance is considerably smaller in streams receiving P inputs. Even in low-P streams, the P-content of both resources and consumers was higher than values reported for temperate streams, suggesting that elevated rates of biological activity in warm tropical streams may alter stoichiometric relationships within a food web. These results show that P-loading in a stream can alter stoichiometric relationships between consumers and their food resources in at least four trophic levels.

       Nutrient retention by streams is an important ecosystem service. Generalizations regarding nutrient retention efficiency of streams in response to nutrient loading have been based on biotic nutrient uptake and short-term (ca. 1-day) nutrient addition experiments. What happens to nutrient retention efficiency when nutrient uptake is primarily abiotic, and nutrient loading occurs over multiple years? Abiotic sorption depends on the degree of sediment saturation and the nutrient concentration in the water column. Where retention is primarily abiotic, uptake efficiency should increase with increasing nutrient concentration, but decrease temporally as sediments become saturated. To test this hypothesis, we analyzed data from an eight-year phosphorus addition (1998-2006) in a 1st-order stream at La Selva Biological Station, Costa Rica. Biweekly measurements of soluble reactive phosphorus (SRP) at three downstream stations resulted in 775 uptake rate calculations. SRP uptake efficiency increased asymptotically with SRP loading due to P-retention by sediments. Uptake efficiency decreased slowly over time, indicated by increasing half-saturation constants over the course of the experiment. Whereas biotic nutrient uptake mechanisms can become instantaneously saturated at relatively low concentrations, our results show that abiotic P uptake does not saturate with increasing SRP concentration but instead saturates over long time periods. 

    Stream ecosystems are influenced by the availability of inorganic nutrients, as well as by the quality of organic matter from the riparian zone. We tested the hypothesis that at high phosphorus (P) concentrations, nitrogen (N) becomes secondarily limiting to leaf decomposition processes in streams draining La Selva Biological Station, Costa Rica. We also predicted that decomposition would occur more rapidly in leaves with a lower carbon to nitrogen ratio (C:N) than in leaves with high C:N. We selected leaves from two common riparian species with high (Ficus insipida) and low (Trema integerrima) C:N. Both species were similar with respect to secondary and structural compounds. Leaves were incubated in three low-order streams with different P concentrations: naturally low (~10 ug SRP L-1); naturally high (>100 ug SRP L-1) due to solute-rich groundwater; and within a whole-stream P enrichment experiment (~200 ug SRP L-1) designed to isolate P effects. To achieve localized N enrichment, fertilizer was added upstream of half of the litter bags. Results suggest P effects; however, our short-term N addition had no apparent effect on leaf decay. In all nutrient treatments, the leaf type with lower C:N ratio decayed most rapidly. Future experiments will examine a larger range in leaf quality and the effects of a whole stream N enrichment.
 
 

    Previous research in streams draining lowland rainforests of La Selva Biological Station, Costa Rica, has shown that algal standing crop accrual in light gaps is significantly less in solute-poor (SRP <10 mg l-1) versus solute-rich (SRP >100 mg l-1) streams that receive inputs of geothermally modified groundwater.  Results from algal surveys in light gaps presented here also indicate that distinct species differences exist between these streams with large Surirella taxa and a previously undescribed Cymbellopsis species dominating high solute streams.  In this study, we also examined algal communities in shaded stream segments, which are more representative of natural conditions in lowland rainforests.  To isolate potential effects of phosphorus from other stream solutes, algal growth and community shifts were investigated in the context of a whole-stream phosphorus enrichment experiment.  Replicate tiles were incubated in shaded sites above and below a phosphoric acid dripper in a solute-poor stream.  No significant differences in chlorophyll a (0.6-1.7 mg m-2), biomass (AFDM 13-21 g m-2) or community composition were found after 35 days.  Results indicate that primary producers are not affected by P in densely shaded streams.  Canopy cover appears to be critical in determining algal response to geothermally introduced solutes in these neotropical streams.

    Previous research in streams draining lowland rainforests of La Selva Biological Station, Costa Rica, has shown that the invertebrate species composition does not differ between solute-poor (soluble reactive phosphorus [SRP] <10 µg  L-1) versus solute-rich (SRP >100 µg  L-1)  streams that receive groundwater enriched with solutes from geothermal activity.  In contrast, algal community analyses shows that solutes play important roles in structuring diatom assemblages in these high- and low-solute streams. Periphyton samples were collected in the wet and dry seasons from epipelic and epilithic substrates in light gaps in streams along a solute gradient.  High- solute streams were dominated by taxa associated with high conductivity [i.e., Seminavis sp. and Bacillaria paxillifer (O.F. Müll.) Hendey].  In contrast, cosmopolitan and acidophilic taxa (i.e., Nupela praecipua Reichardt, Navicula longicephala Hustedt, and Frustulia spp.) were most common in low-solute, poorly buffered streams.  These results suggest that diatom assemblages in high-solute streams are most likely influenced by a suite of solutes, including phosphorus; low-solute streams have diatom assemblages characteristic of poorly buffered, oligotrophic conditions, and, more importantly, low pH.  While these low-solute streams generally have circumneutral pH, diatom response illustrates their poor buffering capacity.

    Soluble reactive phosphorus (SRP) transport /retention was determined in four rain forest streams draining La Selva Biological Station, Costa Rica.  SRP levels can be naturally high due to regional, geothermal groundwater discharge at ambient temperature.  Various short and long-term investigations since 1988 established long-term trends for background SRP and TP (total phosphorus) for three nearby streams with a large range in SRP. Mean SRP  was: 89±53 µg/L for the Salto  (1988-1996), 21±39 µg/L for the tributary, Pantano (1988-1998), and 26± 35  µg/L for the nearby Sabalo.  The Salto in its highlands is low in SRP, until enriched at the slope transition.  SRP concentration was significantly correlated to discharge in the Salto but not in the  Pantano or  Sabalo.  TP was significantly correlated to discharge in all three streams. Since Janurary 1997 Salto discharge and SRP export has been investigated by separate monthly upstream (Salto 60) and downstream sampling (Salto 30 ).   The swamp forest provides a mean long-term contribution to discharge of approximately 36%, but almost 85% of SRP load leaving the watershed.  SRP is not significantly correlated with discharge in the Upper Salto, but is in the Lower Salto. Above the swamp forest SRP concentration was 10-25 mg/L despite discharge, but varied between 40-275 mg/L in the swamp forest depending on discharge.  Pore water analysis using dye tracer rhodamine WT indicated oxidizing conditions, low SRP and high exchange with surface water in Upper Salto.  SRP concentration and conductivity were low. In Lower Salto pore water had low dissolved oxygen, but high SRP and conductivity, and little exchange between surface and pore water.  Enhanced input to high-SRP streams affects biotic metabolism, litter decomposition and secondary production in these tropical environments.

    The relative effects of geothermal solute inputs on diatom community composition and density were evaluated in the context of light limitation in highly shaded neotropical streams.  Geothermally modified groundwater enters streams at La Selva Biological Station in lowland Costa Rica along an elevation gradient.  Diatoms were collected from rocks at 19 sites distributed along nine high and low solute streams.  Light levels were also measured as percent canopy cover.  High solute streams were dominated by six genera (Achnanthes, Bacillaria, Cocconeis, Cymbellopsis, Nitzschia, and Planothidium) while four genera (Eunotia, Frustulia, Hippodonta, and Navicula <10_m in size) were most common in low solute streams.  Canopy cover at both low and high solute sites was >60%.  No significant diatom community differences were found with canopy cover change. Additionally, no differences were found in diatom density between high and low solute streams.  However, community differences did exist, with higher richness (p = 0.0002) diversity (p = 0.02) and evenness (p = 0.05) in high solute streams.  Solute levels, rather than canopy cover, were found to be the driving factors of these differences.
 
 

Top  Home