Cure for Cymophobia

Sarah Zhou Rosengard's stories about water

Month: April, 2014

Slicing up the Amazon

Some things are less obvious about large rivers like the Amazon. In some ways, the water we observe at station Obidos, which is no more than 70 meters deep, is far more complicated than the spacious Southern Ocean, which reaches depths of 4000 meters on average.

Physical oceanographers who specialize in tracing the movement of water beneath the ocean surface will certainly disagree with that statement. Nonetheless, my reasoning stems from how marine chemists, particularly those who care about organic carbon, envision oceans and rivers to work beneath their surface. In general, it is enough for us chemists to assume that most solid particles in the Southern Ocean move vertically by the force of gravity, sinking downward from the surface, from their origins in masses of living algae. Hence, sampling the sinking particle from one location of the Southern Ocean provides a distinct, stand-alone measurement at that particular location of the ocean.

The fact that sediment movement in the Amazon River is water-driven, and therefore predominantly horizontal, changes the game completely. This difference means that estimating the sediment load at one point of the river’s path, such as at Obidos, does not necessarily represent the entire sediment flow across that location; one measurement may not suffice. If one were to cut a slice out of the river at that particular point and view it from its side, water speed and sediment load vary considerably across the entire slice. The result is that, while you might find one sediment concentration and flow rate at the very center top of this slice, i.e., right in the middle of the river, you will find different values 10 feet deeper, or 20 feet away.

We can be sure of this by using the Acoustic Doppler Current Profiler (ADCP). Though small and compact, this instrument can read the river slice by slice. Here, at station Obidos on the Amazon River, we have transited the channel from bank to bank multiple times with the ADCP. With roughly every second in transit, the ADCP measures the range of water speeds as they vary layer by layer, from surface to bottom, generating a vertical profile of current velocity. As it crosses the river, ADCP software glues each vertical profile side by side, creating a mosaic of current velocities at every depth and location of this transect, a slice of the Amazon River. A slightly different crossing would generate yet another slice.


This is how the ADCP looks once it is submerged right under water. It needs to stay here to take measurements.


And we need to stay close enough to read ADCP measurements in real-time.

The learning curve for operating the ADCP and reading the data it collects for us is steep, and I have not yet reached its peak. Nonetheless, I have accumulated enough information so far to understand how much current velocity can very in a given slice of the Amazon River. Though some things about river chemistry are obvious from the bird’s eye view, from the deck of Joao Felipe II, such as its sedimentary composure, it is not obvious that the movement of water along the surface could differ so much at depth. Thankfully, the ADCP extends our ability to perceive these differences.

Because our sediment analysis back in the lab will depend on our understanding of water velocity, data from the ADCP will be particularly useful in making the most accurate estimates from our chemistry. If we take only one measurement from one part of the station Obidos slice, for example, we will be able to understand how representative this one measurement is for the rest of that station.


Clogging the filters

One great advantage of field work in geochemistry is the ability to gain meaningful information from the environment by the simple act of being there. No analytical chemistry necessary. I remember how my advisor Phoebe showed me patches of silvery blue water along the darker back drop of the Southern Ocean surface water during our 2012 cruise on the R/V Revelle. These discolorations indicated dense populations of coccolithophores on ocean surface. Using the naked eye, we were able to hypothesize the chemical features of the water before us.

Several visual observations were similarly immediate and insightful on the Amazon main stem, as well. Upon arriving at the Óbidos site on our first full day, before even taking my first sample – long before actually analyzing it in lab back at home— it was clear that this water was packed with sediment, and by proportion, organic carbon. It was quickly apparent that the amount of carbon I collected per 100 liters at each Fazenda Tanguro stream I visited last November would not measure up to the quantity we could retrieve from filtering the same volume of sediment from the Amazon River waters.


Amazon River sediment clogged on a filter. Though it looks like milk-chocolate, I can assure you it certainly does not taste of it.

As a consequence, the orange murkiness of Amazon River discharge also gave away the unique logistical hurdles of sampling water here in April. On this first day at Obidos, we collected 100 liters of surface discharge near a center point of the river. Using pressure by manual bike pumps, we forced murky orange water through filters designed to collect all particles above 0.22 microns (1 thousandth of a millimeter!) in diameter. The filters clogged after just 1.5 liters. Because a large proportion of the work involved in sample processing during field work is manually exchanging clogged filters for new ones, the greatest challenge ahead was filtering the equivalent of 50 2-liter soda bottles before our trip’s end.

Fortunately, we have an efficient group of scientists on this river boat. João Felipe II is equipped with a diligent team of researchers from the Wood Hole Research Center and Universidade Federal do Oeste do Pará, quick learners of the act of water filtration. With just two bike pumps and three to five scientists, we are able to master a network of 4 filtration cylinders, passing a total of ~9 liters of water through filters at a time.


The filtration team. Even Kaka from Brazil’s 2010 National World Cup team lent a helping hand!

The great sediment concentrations in the river mean that we will go home with more than enough organic material to investigate its origin and transformation in the Amazon River system. We will be able to find specific biological molecules that have been produced by plants, riverine algae, and aquatic bacteria.

More importantly, we will have enough material from these samples to measure the relative proportions of two carbon isotopes in these specific molecules: carbon-12 and carbon-13. Isotopes are atoms of the same element that differ by the number of neutrons in their nucleus. Carbon-12 is the most common carbon isotope in nature; it has six protons and six neutrons inside its atomic nucleus (hence, 12 indicates the sum of protons and neutrons). By contrast, far less common is carbon-13, which is slightly heavier, as it has seven neutrons in its nucleus.

Living things incorporate different proportions of molecules with carbon-12 and carbon-13 when constructing or transforming organic matter. For example, during photosynthesis, plants use less energy in absorbing carbon dioxide with carbon-12 than carbon dioxide with carbon-13. Organic molecules produced by different plant types will demonstrate unique ratios of carbon-12 to carbon-13 based on these preferences.

Resolving the isotope ratios of specific biological molecules in our samples will help us track the history of organic carbon in the river, from its point of production to its journey into the river system and thereafter.