I am keenly interested in how animals interact with each other and their physical environment to control critical ecosystem functions such as carbon flow and nutrient cycling. Because of their size, large spawning densities, and intense physical activity, salmon are the archetypal example of how migratory animals can often control important ecological functions. I am continuing my dissertation work examining the dual role of salmon as vectors of nutrients and agents of disturbance. Most of this work focuses on determining the net effect spawning salmon have on stream ecosystem metabolism (gross primary productivity and ecosystem respiration) and how the geophysical template of streams modulates the amount of physical disturbance by salmon to benthic communities.
Moore et al. (2007) Biotic control of stream fluxes: spawning salmon drive nutrient and matter export. Ecology 88 (5): 1278-1291.
Atmospheric N Deposition to Remote Watersheds
The distinctive signal of human-caused nitrogen dating back more than 100 years appears in 25 lakes in temperate (green circles), alpine (blue) and arctic (red) areas of the Northern Hemisphere. Data from two lakes are shown in comparison to a similar signal revealed in the Greenland Ice Sheet (yellow).
Humans have more than doubled the amount of reactive nitrogen in the biosphere, yet little is known of its accumulation or ecosystem effects outside of heavily populated regions. I am investigating sources of nitrogen to lakes in undisturbed watersheds using lake sediment nitrogen isotopic ratios (δ 15N) as a geochemical tracer. The first of these projects analyzed sediment δ15N back through timein a hierarchical Bayesian models to show a coherent effect of atmospheric N deposition on 24 remote northern hemisphere lakes. A consistent pattern of depleted δ15N in recent sediments indicates addition of atmospheric N sources began around 1895 ± 10 years A.D., concomitant with the beginning of industrialization and emissions of human-caused CO2. In a second project, I am using surface sediment δ15N in lakes with high-quality records of salmon abundance to estimate the total nitrogen loading from 57 remote costal north Pacific watersheds. Salmon have a very distinct nitrogen isotopic signature, and by knowing that signature plus the amount of salmon entering the lake annually, it is possible to back-calculate the amount of nitrogen from watershed sources.
Local-scale Nitrogen Cycling
Measuring N2O emissions from riparian soils in Alaska
At smaller spatial scales, I have used experimental manipulations and comparisons across strong environmental gradients to understand mechanisms controlling how nitrogen is gained or lost from ecosystems. Working in the Matson Lab at Stanford University, we followed an extreme rainfall gradient in Hawai’i to examine changing sources and flux of atmospherically active nitrogen trace gases (NO and N2O) to the atmosphere with soil moisture. In Alaska, I have documented changes to the soil nitrogen cycle with the exclusion of brown bears feeding on salmon from riparian habitats, and showed that the interaction between bears and salmon was critical for generating hotspots of biogeochemical activity. In lake ecosystems of the Pacific Northwest, KathiJo Jankowski, Daniel Schindler and I to use stable isotopes and modeling to quantify N input from N fixation as lakes become increasingly eutrophic.
Jankowski et al. (2012) Assessing non-point source nitrogen loading and nitrogen fixation in lakes using δ15N and nutrient stoichiometry. Limnology & Oceanography 57(3): 671-683 [doi: 10.4319/lo.2012.57.3.06710].
Photos by J. Armstrong
Fisheries of the lower Mekong River Basin are critical to Southeast Asia both as a food resource and a driver of economic growth. Over 50 million people rely on the Mekong River and Tonle Sap Lake fisheries for nutrition, income, and cultural identity. Despite their importance, there is little ecological information about the ecosystem, and many scientific questions about the fishery remain unanswered. A very basic but currently unanswered question is: What are the unique ecological processes that support this unusually large and productive fishery?
The long-term goal of my research in the lower Mekong is to quantify the ecological links within the Mekong-Tonle Sap food web to understand the combination of factors that maintain this highly productive ecosystem. Initially this means firmly establishing the energetic base to the Mekong-Tonle Sap food web. This is a particularly pressing issue for regional food security. Current proposals for hydroelectric dam projects threaten the long-term sustainability of the fishery by potentially removing the annual flood dynamics believed to be important for fish growth and reproduction. Combining stable isotope analysis of consumers and their prey with data-driven ecosystem models of primary productivity, we will be able to better understand which fish species and aquatic communities are most likely to be negatively affected by the removal of seasonal flooding with construction of dams.
Holtgrieve et al. (in review) Ecosystem metabolism and support of freshwater capture fisheries in the Tonle Sap Lake, Cambodia. PLoS ONE
Irvine et al. (2011) Spatial and temporal variability of turbidity, dissolved oxygen, conductivity, temperature, and fluorescence in the lower Mekong River-Tonle Sap system identified using continuous monitoring. Int. J. River Basin Management 9(2): 151–168.
Aquatic Ecosystem Metabolism
Bayesian Modeling of Ecosystem Metabolism Using Diel Oxygen Data
Quantifying rates of primary production and respiration (ecosystem metabolism) is critical to our understanding of energy flows and nutrient cycling, sources and fate of carbon, ecosystem trophic state, and food web dynamics in aquatic ecosystems. I am actively developing models of aquatic ecosystem metabolism, including Bayesian methods to estimate metabolic parameters (BaMM, for Bayesian Metabolic Model). These models use day-night (diel) cycles of O2 concentration, water temperature, and irradiance data to estimate probability distributions of ecosystem metabolic rates such as gross primary productivity, community respiration, and air-water gas exchange.
Holtgrieve et al. (2010) Simultaneous quantification of aquatic ecosystem metabolism and re-aeration using a Bayesian statistical model of oxygen dynamics. Limnology and Oceanography 55 (3): 1047–1063.
Parameterizing Air-Water Gas Exchange
Aquatic ecosystems accumulate material from their surrounding watersheds and as a result play a critical role in the global carbon cycle. One of the most challenging aspects in determining fluxes of carbon—or other biogenic gasses such as oxygen—through aquatic ecosystems is understanding the rate in which these gases exchange with the atmosphere and the physical factors that control this process. The BaMM model provides a new way to estimate the gas transfer velocity directly from dissolved oxygen data by exploiting the large amount of information contained in continuous data. I am also working on a meta-analysis of dual-tracer gas-transfer data to better understand uncertainties in CO2 uptake by the oceans as well as collaborative projects to determine biogeophysical drivers of gas transfer in streams and lakes.
Alin et al. (2011) Physical controls on carbon dioxide transfer velocity and flux in low-gradient river systems and implications for regional carbon budgets. J. Geophysical Research, 116, G01009, doi:10.1029/2010JG001398.