For my doctoral studies, I studied the impact of bedrock landslides on chemical weathering in fast-eroding mountain belts. There were a number of hypotheses that my colleagues and I tested over the course of this project, but initially it is useful to understand a little about the broader background and context of this topic.
Chemical weathering is an important process that occurs across the vast majority of earth’s surface; where water and rock interact, there is potential for some of the rock to be dissolved and deposited elsewhere, whether into the ocean or dry basins. This forms an important component of the cycling of elements across earth’s crust, and in part affects nutrient fluxes into the ocean. Crucially, when rocks containing silicate minerals are dissolved, they can draw carbon dioxide out of the atmosphere. Since the rate at which these reactions occur is a function of temperature, this can act as a long-term control on the amount of carbon dioxide in the atmosphere (this is the basis of the suggestions for an ‘enhanced weathering’ style of geo-engineering).
Large bedrock landslides are well known natural hazards; they represent a danger to anyone living in steep mountainous regions, or where earthquakes are also a risk. It is reasonable to ask why these might be of interest from a chemical weathering perspective, though. We hypothesise that they might have two main effects; first, the extensive fragmentation and shattering of rock during a landslide presents a huge amount more surface area for percolating rainwater to interact and react with, which could increase dissolution. Secondly, landslides can excavate rock deep within the bedrock. Since most rocks contain a variety of minerals that dissolve at different rates, this means that exposing a large amount of fresh bedrock could provide a surplus of the most highly reactive minerals, which would then dissolve more quickly in the landslide deposit. With these aspects in mind, we aimed to test a number of hypotheses:
1. Does the presence of landslides affect weathering, both at the local scale and catchment scale?
The primary target of this study is to establish if landslides act as local foci for rapid chemical weathering, and if they do, to what extent this affects catchment scale solute budgets where landsliding is extensive. If landslides are significant on a catchment scale, is it possible to observe a transition from a landscape where soil production dominates weathering to one where landslides are more important?
2. Does weathering associated with landslides differ from other locations of weathering in the landscape? Why?
If landslides can act as sites of accelerated weathering, the next question is why this is the case. Which physical characteristics of landslides are most important in setting the weathering? Do landslides have an excess of soluble or highly labile material in the deposit, or does the weathering result from an elevated acid levels in the deposits? Does this result in different ratios of dissolved elements from landslides in comparison to more stable parts of the landscape, and does this affect proxy based measurements of weathering?
3. Can we quantify the sources of acid and the substrates weathered in landslides?
How does the physical process of landslides affect the proportion of acid derived from various sources? Can macerated organic matter or trace sulphides incorporated into the deposits be effectively oxidised to provide a source of acid, or is atmospherically sourced carbonic acid the key weathering agent? What are the fractions of silicate and carbonate rock that are weathered in comparison with their abundance in the mobilised lithology?
4. How does this style of weathering affect the carbon cycle?
Quantification of the sources of acid and the substrates that are dissolved in weathering associated with landsliding should provide an estimate of the impact of such weathering on the sequestration of carbon dioxide. Does weathering driven by landsliding act to link erosion and climate through this sequestration?
In the thesis, these questions were addressed in four chapters, which were themselves mostly based on a single research paper, either published or in preparation.
Q1: We found that in the Western Southern Alps of New Zealand, landslides are important locations for chemical weathering; in landslide deposits, the weathering seems to be faster than other parts of the landscape. We also found that this effect is of primary importance at a river-basin scale, and can contribute up to half of the dissolved material in the rivers draining this rapidly eroding region.
This study was published in Nature Geoscience: doi:10.1038/ngeo2600
Q2: In the southern mountains of Taiwan, we explored what causes landslides to have such an impact on chemical weathering. There, the landslides expose small amounts of highly reactive minerals which dissolve quickly, and thus accelerate the rate of weathering. This was published in Earth Surface Dynamics: doi:10.5194/esurf-4-727-2016
Using more detailed analysis of our previous results from New Zealand, we also showed that the rapid dissolution of such small mineral phases can adversely affect the reliability of commonly used techniques (‘proxies’) to assess rates of chemical weathering over very long time periods. This study has just been accepted for publication in Geochimica Cosmochimica Acta.
Q3 & Q4: We used a large compilation of data from both Taiwan and New Zealand to show that rapid weathering of rock in landslides is partly due to an increase in the acidity of the fluid percolating through the landslides. This acid is importantly derived from the breakdown of organic material within the landslide itself; in Taiwan this is supplemented by sulphuric acid from the dissolution of minor amounts of the mineral pyrite in the landslides.
In Taiwan, the combination of acid derived from sulphide minerals and dissolution of primarily carbonate-rich rock means that the net effect of the weathering actually increases atmospheric carbon dioxide, which further throws into question the importance of high erosion rates as a negative feedback on atmospheric carbon dioxide – which has been a topic of study for many years.
This larger, final paper, will soon be submitted for consideration in the coming months.
I personally found this work fascinating, especially since there are a number of further questions for research that could be pursued that stem out of the work. What is the importance of biology (and particularly microbes) in setting this style of weathering? What impact do landslides have at a global scale? Can the landslides triggered by earthquakes have a measurable effect on mountain-belt scale chemistry?
Although I have been focused on other topics in the last few months, I remain keenly interested in this subject, so if this opens up questions, please do let me know via the contact page.
Robert Emberson 