#05 2021

A quest for soil CO2 flux data: lessons learned from 33 months of field data collection before and during the pandemic

By Lindokuhle Dlamini, PhD Student with the University of the Free State (South Africa) and the University of Burgundy (Dijon France), hosted by the SAEON Grasslands-Forests-Wetlands Node

There is a global movement to plant trees to mitigate climate change (including the Bonn Challenge, AFR 100 and UN Trillion Tree campaign). This is driven by international agendas that negate evidence-based science from open ecosystems in Africa. Planting trees in well-managed open systems like African grasslands will not solve the climate crisis; it might create even more problems. These intact, well-managed and diverse ecosystems are vital for assisting climate mitigation efforts. They need to be managed in conjunction with reducing fossil fuel emissions (not instead of).

The United Nations recently launched the Decade on Ecosystem Restoration to prevent, halt and reverse the degradation of ecosystems worldwide. While the initiative is catalysing restoration efforts across the globe, care should be taken that these efforts are ecosystem appropriate.

South Africa’s grass-dominated ecosystems include fire-climax grassland in many of the key strategic water source areas, such as the Drakensberg Mountains. Until now there has been a paucity of in situ data exploring the carbon dynamics of these systems. We need such data to be able to assess the role of different ecosystems in the carbon cycle and the potential impacts of land-use change, for example from grasslands to exotic afforestation or indigenous woody dominated systems, to provide evidence-based information on the consequences of such changes.

Victory hike – Lindo walking towards the Cathedral Peak hiker’s hut after finishing his data collection for the last time.

For my PhD, I set out to provide much needed data points in southern hemisphere grassy systems on temporal and spatial soil CO2 flux. This work will provide detailed data on soil carbon composition and dynamics under several experimental treatments, elucidate on mechanisms driving soil carbon storage in grassy systems, as well as guide restoration work and contribute to climate change adaptation strategy.

During this time, I became involved with the SAEON Graduate Student Network and came to realise the importance of sharing my research experiences with fellow students. With the fieldwork component of my work now complete, I reflect on the journey thus far and offer some teaser results.

To do science is to search for patterns and understanding, not simply to accumulate facts. For me this meant month-to-month visits to remote locations in the most beautiful mountains in KwaZulu-Natal for almost three years. Conducting research in such locations, however beautiful, can be challenging if you are going at it alone and the PhD process itself can be lonely, hard and emotionally draining. Add to this the misery, uncertainty and rules that came with the pandemic and the challenges can seem insurmountable.

Benefits of the SAEON platform 

I have learned that a strong support system, personal passion and commitment are crucial in undertaking a PhD and doing long-term field-based research. SAEON is an enabler, and it is indeed a privilege to have had so much support and equipment to undertake such exciting research in some of the most remote, yet breathtaking, montane grasslands. By being part of a team and harnessing the benefits of the SAEON platform, I was able to enjoy my field research and pause from time to time to take in the stunning scenery.

My PhD work (see here) focuses on soil carbon dynamics of fire-prone grasslands in Cathedral Peak. I am interested in two aspects: 1) Carbon release (soil respiration or soil carbon dioxide [CO2] flux using a monthly manual method and a continuous long-term LI-8100A Automated Soil Gas Flux System owned by SAEON); and 2) Carbon storage through quantifying current carbon stocks in an entire catchment at different soil depths and within different aggregate sizes, as well as carbon isotopic signatures to understand the contribution from C3 trees and C4 grasses. My work thus required monthly week-long visits to sites to take measurements, check equipment and get a long-term time series of repeated measurements through the seasons.       

A pioneer forest site in a fire-excluded catchment with manual chamber soil CO2 measurement and soil microstation.

An adjacent grassland site in a fire-excluded catchment with manual chamber soil CO2 measurement and soil microstation.

The calm before the storm: sampling in beautiful mountains with unpredictable weather conditions  

I am passionate about the kind of research that I do, and I often start with the end goal in mind. Sampling in the Drakensberg Mountains is an amazing experience and for the first few months, fieldwork felt like a holiday. Indeed, being exposed to new equipment and a new kind of science was among the most interesting experiences.

Coming from a grassland and restoration ecology background, with limited knowledge of soil CO2 flux measurement and equipment, SAEON and the mountains allowed me to explore and learn more about in-situ data measurement equipment and multidisciplinary research. The mountains made clear the interconnectivity and dynamic nature of environmental research – from the atmosphere to the vegetation to the soils to the water. Working in these majestic mountains is like working in a multidisciplinary living laboratory.

But I did encounter some challenges. The beauty of the Drakensberg landscapes can go unnoticed if your goal is to finish a time-dependent four-hour (sometimes six-hour) soil gas sampling per site. This can be especially difficult in summer knowing that weather conditions can change within a matter of minutes.

While it was possible to work in the rain and during gale-force winds, thunderstorms would end field sampling because of the possibility of lightning strikes. We experienced all these conditions but were able to escape thunderstorms and run to safety just in time, sometimes halting field sampling an hour after commencing it.

Fluctuating temperatures would also kill the batteries of a soil MicroStation, sometimes on a monthly basis during the winter months. Aside from weather conditions, porcupines would dig out sampling collars affecting the experiment, while other small animals (possibly rodents) would chew soil MicroStation cables.

All these difficulties taught me a lot about planning ahead and I quickly learned how to fix the equipment. These are lessons I would not have learned if weather conditions were smooth and there were no pesky porcupines on site!

CO2 flux changes over time in a fire-excluded catchment showing differences between a pioneer forest site with no fire and adjacent grassland site with fire.

Air temperature changes over time in a fire-excluded catchment, showing differences between a pioneer forest site with no fire and adjacent grassland site with fire.

Gaining life skills, not just science skills  

Adding to the weather conditions, the roads are not for the faint-hearted and require careful negotiation. We were fortunate that SAEON sent us for off-road 4×4 training. The skills we learned came in very handy on our field trips. The drive up the steep and bumpy Mike’s Pass became a smooth, often unnoticeable, drive as we were always looking forward to getting to the sampling sites.

I use “we” or “us” because SAEON standard operating procedures require that no one should drive or be allowed to work alone in the field. I always worked with my fellow PhD student Rowena Harrison. We planned our trips to coincide with the SAEON technicians’ routine field visits and were thus able to learn more about teamwork under the leadership of node technician Kent Lawrence.

A fundamental lesson we learned is the importance of good and regular communication. We also learned that our successes are interdependent, which sometimes means putting what works for the team above personal preferences.

Beating the pandemic  

The pandemic did not spare us. After the first hard lockdown in 2020 we were not allowed to go out into the field. However, we resumed fieldwork in May 2020 because SAEON rapidly put standard operating procedures in place to ensure we keep each other safe in the field and yet still work effectively.

This was a huge adjustment but going to the mountains became the best escape after we had been living in isolation indoors because of the lockdown. Yet again the mountains saved us from the pandemic – that one-week fieldwork in a month offered a much-needed escape from confinement.

Soil temperature changes over time in a fire-excluded catchment, showing differences between a pioneer forest site with no fire and adjacent grassland site with fire.

Relationship between soil temperature and CO2 released from the soil between a pioneer forest and adjacent grassland site within a fire-excluded catchment, Cathedral Peak, Drakensberg.

Discovering new passions and effective coping strategies  

Weather conditions in Cathedral Peak are unpredictable, but checking the weather forecast on the day before the trip often proved to be a reliable barometer. Carrying a toolbox with spares and batteries became the norm, as well as doing fieldwork wearing a mask.

The uncertainty outlined above often led to fieldwork fatigue, but the support system SAEON provides and my newly developed love for landscape photography and astrophotography alleviated some of this anxiety. The mountains stimulated my love for photography, which is why fieldwork remained interesting regardless of some of the difficulties.

Preliminary results 

When one tackles intensive fieldwork over a long period of time, it is exciting to see how the data accumulate month after month. At this stage I am exploring the different data sets and preparing for analysis. For the carbon release component, I am analysing manually collected month-to-month data as well as data from the continuous long-term LI-8100A Automated Soil Gas Flux System (the only one of its kind to take continuous measurements in a natural system in South Africa that we know of).

In one of the catchments – a fire-excluded catchment dominated by woody plant species, especially Leucosidea sericea – I am comparing CO2 released from the soil in two sampling sites:

  • Pioneer forest: where fire is excluded and hence does not burn often.
  • Adjacent grassland: an outlying area within the same catchment that experiences accidental fires every two to three years.

Initial results from this catchment clearly show that pioneer forest displays a different pattern in the CO2 released compared with the adjacent grasslands and that it releases more CO2 from the soil with an increase in temperature, with greater sensitivity to soil temperature even though it remains relatively cool compared with adjacent grasslands. Minor changes in temperature have a huge impact on the amount of CO2 released.

One of my other study catchments is catchment 6, a “pristine” grassland catchment where the Licor is deployed. I have developed R code for cleaning this enormous data set which takes flux readings from six chambers every hour. I will be analysing these data to understand seasonal changes in soil CO2 flux and the impact of climatic variability of these fluxes. I will also compare these data with other similar systems in the world from the global database.

The CO2 flux data from the continuous long-term LI-8100A Automated Soil Gas Flux System showing data from late 2017 to early 2021. There are gaps in the data because the system was either removed for calibration and maintenance in the United States of America or removed to protect it from fire damage.

Lindo at the Cathedral Peak hiker’s hut after completing his data collection.

Last day of fieldwork  

In order to enjoy science and fieldwork, you must have fun while doing it. I enjoy being outdoors and that is the one thing I will miss the most. To celebrate my fieldwork achievement and say a special goodbye to the mountains, I went on a long hike to the hiker’s hut to enjoy the beautiful scenery for the last time before the start of a long data-analysis and thesis-writing process.

No doubt when I need inspiration, I will sneak out on one of the SAEON field trips and spend time contemplating on a rock, for it is by being within the systems that we are researching where discovery is most likely to take place and where we can best interpret the meaning of the numbers on our screens.


I would like to extend a special thanks to Sue Janse van Rensburg, coordinator of the Grasslands-Forests-Wetlands Node, as well as Rowena, Kent and Busi Mdunge, along with the entire node team for their endless help and support during data collection for my PhD. I would like to acknowledge EFTEON for funding and the French team for guidance, analysis of samples and supervision.

A special thanks to Dr Gregor Feig (manager of EFTEON) and Dr Elmarie Kotzé (Senior Lecturer at the University of the Free State, Department of Soil Crop and Climate Sciences) for their mentorship and supervision.