Nitrogen dynamics and retention in the river network of a tropical forest, Luquillo Mountains, Puerto Rico

This dissertation identifies gaps in the scientific understanding of nutrient cycling, particularly nitrogen (N) cycling, in streams and riparian zones of tropical montane forests, and addresses several of those gaps with original field-based research using study watersheds in the Luquillo Mountains of Puerto Rico as the model system. The Luquillo Mountains have features typical of mature montane tropical forests, such as high background N concentrations in streams and groundwater relative to streams in other biomes. As a USDA Forest Service Experimental Forest, the Luquillo Mountains are accessible to researchers and have abundant monitoring and experimental datasets from which to build hypotheses and experimental approaches. Chapter 1 is a review of the state of the literature on biological response to nutrients, particularly N, in streams of the Luquillo Mountains. This chapter also includes a gap analysis of research questions that are of greatest importance to the environmental regulatory and management community, that have not yet been addressed. Chapter 2 looks in-depth at ammonium (NH4+) cycling in headwater streams. Headwater streams in tropical forests are typically light- and organic matter-limited in their demand for nutrients. Ammonium can serve as both a nutrient and an energy source, leading to a hypothesized high demand that can be compared across streams using ambient uptake velocity (vf). This study experimentally enriched headwater streams with transient NH4+ pulses to determine NH4+ demand and mechanisms for uptake. Ambient vf ranged from 0 to 2.9 mm min-1, lower than other tropical streams and streams in other biomes in the literature. Though demand was relatively low, areal uptake rate in the streambed was high due to high background NH4+ concentrations. When compared with streams in other geological regions of the Luquillo Mountains, the streams in this study stand out for their sandy substrate and their low phosphorus concentrations, suggesting that NH4+ removal pathways may be limited by nutrients or habitat for NH4+-oxidizing microorganisms. Chapter 3 focuses on N cycling in riparian zones. Riparian zones are widely understood as nitrogen N cycling hotspots, but significant gaps remain in our understanding of the complex biogeochemistry of NH4+ transformations in riparian groundwater. Tropical forest watersheds in particular have distinctive N biogeochemistry that is still poorly understood. This study was the first to examine in-situ NH4+ cycling in a riparian aquifer, using push-pull tests PPTs to experimentally enrich groundwater with NH4+ and trace NH4+ removal from solution, and the first to apply the Damkohler number to riparian NH4+ dynamics to measure the balance between residence time and reaction rate. The rate constant k for NH4+ retention during the five PPTs ranged from 0.13 to 0.68 hr-1; the residence time ranged from 27 to 512 days; and the Damköhler number ranged from 72 to 11620, indicating that nearly complete removal of added NH4+ would occur over transport from the PPT well to the stream. Low dissolved oxygen availability and lack of net nitrate production indicate that nitrification was not the dominant pathway of NH4+ removal. Iron was abundant in surface riparian soils in the form of HCl-extractable Fe(III), and declined to near zero at the depth of the PPTs accompanied by abundant HCl-extractable Fe(II), suggesting that Fe(II) production and NH4+ oxidation could be coupled. Sorption-desporption reactions and the associated equilibrium potentially explain why high background NH4+ concentrations persisted before and after the PPTs, though we expect that NH4+ pulse experiments like those conducted here saturate abiotic storage within soil and provide a determination of biotic NH4+ removal