The surface of Earth is being reshaped constantly. Mountainous uplands are broken down by water and wind producing sediment that is moved by rivers to lowlands. Some of this sediment is deposited along the way, some is delivered to the coast and continental shelf, and some makes its way to the ultimate sink, the deep sea. This transfer of material across the Earth's surface creates the landscapes we inhabit.

However, the reshaping of the planet's surface varies geographically and varies through time. How long does it take for sediment to be eroded from mountains? How long does it take for that sediment to be transported from mountains to the coast? How long does it take for it to be moved from the coast to the deep sea? What are the pathways the sediment takes from source to sink? How do these attributes differ from system to system, or at different times in Earth history?

Answering these questions has implications for understanding how other materials -- pollutants and carbon, for example -- are transported and distributed across the Earth's surface. Importantly, insights about the controls on sediment deposition through space and time are critical for improving our ability to read and interpret the geologic record -- the archive of Earth's history.

One way to try and answer these questions is to determine the sediment budget of a system. Let's consider sediment the "currency" of the Earth's surface -- it can be withdrawn (eroded), transferred from one account to another (transported), and, well, deposited. If we could track the sources, movement, and destinations we'd be much closer to answering the questions I posed above. Unlike actual currency, however, we can't track the movement of sediment with such accuracy. Furthermore, we want to determine sediment budgets for time periods before our modern world to gain information about how the Earth's surface responds to global change at longer timescales (centuries to millennia).

In other words, how do rates of sediment production in the erosional part of a system compare to rates of sediment accumulation in the depositional part of the system at timescales of thousands of years?

A new paper I'm a co-author on, coming out in the July issue of Geology and online early here, summarizes the results of research asking this very question.

What Did We Do?

Determining rates of sediment accumulation is relatively straightforward (once you have the data). The volume of sediment in the deep-sea fan was determined from mapping on publicly available seismic-reflection data. These volumes were then tied to existing cores in the depositional system that had radiocarbon ages, which provided the constraint on the timing and, thus, the depositional rates.

Calculating rates of erosion, especially at long timescales, is a bit trickier. In this case we used the abundance of an isotope of beryllium (10Be) that is produced in rocks at the surface of the Earth from cosmic radiation. Essentially, the slower a landscape is eroding the larger the abundance of these cosmogenic nuclides, as they're called. The faster a landscape is eroding, the smaller the abundance. By measuring the abundance in river sands collected near the outlet of a drainage basin you can then calculate an average erosion rate for that drainage basin that is valid for timescales of thousands of years * .

Actual sedimentary systems are more complicated than the cartoon I drew above. Commonly, there are multiple catchments that might feed a single depositional area and sediment transport laterally along the coast needs to be considered. Additionally, we were interested in how the sediment budget -- the balance of erosion and deposition -- changed with the significant sea-level change since the last ice age 18,000 years ago.

We chose to conduct this study using systems in southern California because of the exceptional context from previous research. Like an experiment, we wanted to know as much as possible about the boundary conditions and cause-and-effect relationships. There will always be uncertainty when using nature's experiments to ask questions about how the Earth works, but here we think the existing knowledge about these systems reduces that uncertainty.

What Did We Find Out?

The figure below is from our paper and summarizes the main findings. The left part of the figure depicts the systems when sea level was ~130 m lower than at present (during the last ice age when water was locked up in continental ice sheets). The right part of the figure shows the condition from ~15,000 years ago to present when sea level rose.

Although the different sea-level stands influenced the pathways and the ultimate site of deposition for the sediment, the mapping and sampling for the study took all this into account. In other words, we have accounted for nearly all the sources and sinks for this sedimentary system even as sea level changed.

The graphs at the bottom of the figure above summarizes the rates of deposition and rates of erosion (or denudation). In the low sea-level state (at left) deposition and erosion is the same. That is, at these timescales all the sediment that is eroded from land is making its way to the deep-sea fan. When sea level is rising and at its current high position (the graph at right) note that depositional rates are a bit higher than the erosion rates. There is more sediment than can be accounted for -- there is a surplus of sediment in this budget. We think that erosion of the coast during sea-level rise could be contributing this "missing" sediment.

But, even in the high sea level condition, the rates are broadly similar (there isn't an order of magnitude imbalance, for example). For these relatively small systems the sediment that is produced from erosion of these coastal mountains is transferred to the depositional parts of the system over thousands of years. This is intuitive because there are few spots along the pathway on land for sediment to be "stored" for long periods -- these rivers and streams come out of the mountains right at the coast. In much larger systems, however, there is ample space (in the floodplains of rivers, for example) for sediment to be stored for thousands or even millions of years. In other words, to accurately evaluate the long-term sediment budget for those larger systems you'd have to account for that deposition on land, at the coast, and in the deep sea.

What I find most interesting is what all this means for examining the stratigraphic record. Although these erosion and deposition rates are at timescales much longer than human observation they are still very short compared to the geologic record. As we go back further in geologic time, we lose the ability to determine process rates at this resolution. Also, by its very nature, mountainous uplands are not preserved -- they are completely eroded away. Can we reconstruct those ancient landscapes that are long gone by examining the stratigraphy it produced?

Studies like the one I've highlighted here are a bridge to understanding landscapes in deep time and will help us unravel the controls of Earth surface systems. There is still much work to do, it's an exciting time to be thinking about these problems.

*There is, obviously, a lot more detail to this method of calculating denudation/erosion rates that I didn't have space here to cover. I highly recommend this 2006 paper by von Blanckenburg for those wanting to dive into the theory and application. For a less technical description of the method, this article and accompanying video is a superb introduction to the subject.


Covault, J.A., Romans, B.W., Graham, S.A., Fildani, A., & Hilley, G.E. (2011). Terrestrial source to deep-sea sink sediment budgets at high and low sea levels: Insights from tectonically active southern California Geology, 39 , 619-622: 10.1130/G31801.1


1. Simplified sketch of coupled erosional-depositional sedimentary systems across a continental margin. Drawn by the author.

2. Figure 2 from Covault et al., 2011, Terrestrial source to deep-sea sink sediment budgets at high and low sea levels: Insights from tectonically active southern California Geology, 39, 619-622: 10.1130/G31801.1

About the Author: Brian Romans is an assistant professor in the Department of Geosciences at Virginia Tech where he specializes in sedimentary and marine geology. Brian shares his passion for Earth science through words and images on his blog Clastic Detritus at Wired and on Twitter as @clasticdetritus.

Cross-posted at Clastic Detritus.

The views expressed are those of the author and are not necessarily those of Scientific American.