Amazon Research Objectives:

   

    CAMREX.   We have a long history of working on the biogeochemistry of the Amazon. Following a pilot "cruise" of the UNOLS vessel Alpha Helix, in 1977, the NSF, NASA, and Brazilian- supported project CAMREX (Carbon in the AMazon River Experiment) began, in 1982. The overall perspective in CAMREX has been that the Amazon is a test case for developing extendable models of how hydrologic and biogeochemical cycles are coupled at regional to continental scales in the humid tropics. Our studies serve the dual purposes of gaining a broad mechanistic understanding and of establishing data baselines needed to assess anthropogenic perturbations to these globally critical and ecologically complex systems. As documented in over 150 publications, the CAMREX dataset represents a time series unique in its length and detail for very large river systems. The focus over the last several years has been to examine the sequence of processes involved in the dynamics of carbon metabolism, resulting in high outgassing. Richey et al. (2002) showed that the outgassing from the aquatic systems of the Amazon are roughly equivalent to the carbon sequestered on land, and over 10x the carbon exported to the sea. Papers have addressed modeling and system integration (Richey et al. 2004; Richey 2004, 2005), biogeochemistry within geospatial frameworks (Ballester et al. 2003, 2005; Bernardes et al. 2004; Mayorga et al. 2005a, Logsdon et al. 2005; Krusche et al. 2005), hydrologic models (Victoria et al. 2007), gas exchange (Rasera et al. 2008; Alin et al.2011), basin scale tracers of sorption and metabolic properties (Martinelli et al. 2003; Mayorga et al. 2005b; Aufdenkampe et al. 2007; Remington et al. 2007; Souza et al. 2008; Ellis et al. in press), and chemical properties (Dickens et al. 2007, Tumang et al. 2007). Richey et al. (2010, 2011) summarized the current understanding of carbon in Amazon rivers. 

     A signature of CAMREX is that is has been a joint, collaborative effort of the UW and Brazilian institutions, in particular CENA (Centro de Energia Nuclear na Agricultura, Piracicaba SP) and INPA (Instituto Nacional de Pesquisas da Amazonia, Manaus).  As part of LBA, we established the education and sampling network the Rede Beija Rio (RBR), supported subsequently by the Brazilian government (FAPESP, CNPq). The RBR serves the dual purpose of enabling higher frequency (and less expensive) sampling than possible by sending teams from São Paulo or Seattle, and especially serves as our primary vehicle for training and capacity building. The sites are organized as transects, from the Amazon mainstem and major tributary mouths (the initial CAMREX scheme), to primary tributaries, then sub-tributaries, and ultimately streams The RBR is made up of nodes distributed across the Amazon, where each node is occupied by a researcher or a team of researchers from that site-typically a professor, graduate and/or undergraduate students from local institutions, working from a coordinated sampling plan. Over 40 Brazilian students have participated in the project since its beginning, earning Masters and PhDs. Most of these students have been from the Amazon, a region markedly under-represented in technical training in the country.   

      The Current Project: RioROCA Just as the role of gas evasion has classically been overlooked in the carbon balance of fluvial systems, we have also neglected to fully examine the sequence of processes that occur when freshwater meets the sea. Most, if not all, characterizations of the mass flux of material from rivers to the ocean have been performed at the downstream-most gauging station on that river. Such stations are typically located a considerable distance from the actual confluence of the river with the sea, usually above tidal influences. In the case of the Amazon River, for example, the defining gauge is located at Óbidos, some 900 km from the sea (Fig. 1). In the reach below Óbidos, three major tributaries  plus many smaller tributaries and local flood plain drainages, add ~40,000 m³/s, or an additional 20% to the river's total annual discharge (about two Mississippi Rivers' worth, with very different chemistry, than upstream). The Amazon plume itself, fed by river nutrient inputs from P, Si, and Fe to the outer plume and N to the inner plume, appears to reverse the normal tropical surface ocean condition of outgassing, and leads to CO₂ uptake and sequestration in areas that would otherwise be outgassing (Takahashi et al., 2002, Cooley et al., 2007, Subramaniam et al., 2008). A comparison of fluxes between Óbidos and offshore sediment accumulation by Nittrouer et al. (1995) suggested that roughly a third of the sediment from upstream is captured somewhere in the tidal river. Richey (2011) suggested that the inclusion of the terrestrial-driven marine fixation would essentially double the "effective" impact of the river on the ocean since river OC export is roughly equivalent to river-supported new production in the plume. But our knowledge of export is very low. The actual calculation of the plume fluxes is dependent on knowing what the river end member dissolved inorganic carbon signal is, which has never been actually measured. 

     ROCA (River-Ocean Continuum of the Amazon) is a project designed  to establish the overall chemical, biological, and physical "continuum" of the Amazon River system, from Óbidos (the last routine measuring station, 900 km inland), to the outer marine plume (1000 km to sea). The intention is Improving through models and observations our predictive capabilities of the microbial ecology and biogeochemistry of the Amazon continuum. It is a multi-institution project, based at the University of Georgia (P. Yager, PI), funded by the Gordon and Betty Moore Foundation. Briefly, the process of ROCA is to bring river scientists and oceanographers together to fill the gap between river and ocean ecosystems, and extend the tropical river continuum through the lower reach to the open ocean. The main objectives of this program are to improve our understanding of carbon exchange between the atmosphere and tropical river continuums (including the plumes), starting with the Amazon River, focusing on the lower reach, nearshore, and offshore tropical Atlantic, thus enhancing predictive capabilities under differing climate change scenarios. As the common driver of material processing down the continuum of the Amazon, from land to sea, is the microbial community, attention extends beyond carbon and nutrients to include the "genomic" structure and succession of the bacteria and phytoplankton communities.

     The targeted outcome of ROCA is the "Enhanced predictive capabilities regarding the interplay between marine microbial communities, biogeochemical cycling and carbon sequestration in a major river plume environment and to understand the sensitivity of these interactions to environmental change." ROCA consists of a set of Modules: Module 1. "Bridging the Gap between the Amazon River Ecosystem and the Ocean," Module 2. "Metagenomics and metatranscriptomics: community succession and gene expression through the river-ocean continuum," Module 3." Enhancing the predicative capacity of ecosystem models through the incorporation of metagenomics and metatranscriptomics data.

     RioRoca represents the overall team focusing on the river portion of the project, from Obidos to the sea.  We have developed a protocol which allows the (first-ever) measurements of the discharge and chemistry of the lowermost Amazon, coupling measurements at Óbidos (Obi) and the  Rio Tapajós to the total flux to the sea. The river mouth is divided into 3 channels (each ~11 km wide), 2 by Macapá  (to the north of Marajo Island), with MacN (to the north of a mid channel island), and MacS ( to the south), and Bel, above Belém (to the south of Marajo).   Discharge is measured by continuously collecting ADCP profiles across each channel, over the complete 13-hour tidal cycle.  The following day, each channel is sampled at equidistant-points in the right, mid, and left channels, for all chemical parameters at surface and mid-depth by in-situ pumping. River cruises were conducted in September ‘10 (Falling water), December ‘10 (low/earliest rising water), May'11 (high water), and September ‘11 (falling, more complete sampling).

     Initial results from the first three cruises have been reported at the San Francisco AGU 2011 meeting. Richey et al.