Exhuming the Climate of Indonesia: 
The Towuti Drilling Project

Satrio Wicaksono ( S W )
in conversation with Anna-Sophie Springer ( A S )
& Etienne Turpin ( E T )
Published 21 June 2016

The diverse natural landscapes of central Indonesia inspired Alfred Russel Wallace to conceive of natural selection and biogeography theories, yet the region’s long-term climatic and environmental histories are virtually unknown. From May to July 2015, the Towuti Drilling Project (TDP)—the first lake drilling project anywhere in Southeast Asia—conducted an international scientific expedition to drill into the sediment underneath Lake Towuti in South Sulawesi. The obtained drill cores will be used to reconstruct the climatic and environmental evolution of central Indonesia during the past 800 millennia through biogeochemical and physical analyses. Simultaneously, this research highlights the natural processes that helped generate the landscapes Wallace saw during his own expedition between 1854 and 1862. Anna-Sophie first met Satrio, then Project Coordinator of the TDP and a Ph.D. student at Brown University, during the 12th Northeastern Conference on Indonesia at Cornell University in the fall of 2014. After an ongoing email correspondence about the unique research, and a failed visit to Lake Towuti in June 2015, we had the chance to speak about his research on the occasion of 125,660 Specimens of Natural History.

A S
Can you explain the origins of the Towuti Drilling Project? What began your research in this region of Sulawesi? Why is this area important? And what is the overall aim of the project?

S W
The Indonesian archipelago and the seas surrounding it are a major source of global water vapor and heat, and therefore play a crucial role in the global climate system. A better understanding of the convection and precipitation mechanisms around the archipelago over time will not only give us a better idea of what Indonesia’s precipitation will look like in the future but also give us deeper insight into the past, present, and future of the global climate system. Unfortunately, there are only limited climate data available from Indonesia. Based on our previous research, Lake Towuti, located at the heart of the Indonesian archipelago, contains hundreds of meters of sediment that can be used to reconstruct climatic and environmental changes during the last circa 800,000 years, the only known such record from the region. The Towuti Drilling Project is the first scientific continental drilling project in Southeast Asia, and from a paleoclimatic perspective, its location is just perfect. My Ph.D. advisor at Brown, Dr. James Russell, first envisioned the TDP about a decade ago, when he started working in the Indonesian region. As an expert on both past tropical climates and lake ecosystems, he is interested in reconstructing and deciphering climate and environmental information contained within layers of mud buried underneath tropical lakes. Together with Prof. Satria Bijaksana from Institut Teknologi Bandung, James started some preliminary research at Towuti and surrounding lakes in 2007. After multiple fieldwork sessions and analyses of preliminary data, the two of them, along with their collaborators, began writing a proposal for the drilling project in 2012.

E T
You mentioned previously that you have done research in the region prior to this project—can you tell us more about how you prepared?

S W
We conducted seismic expeditions in 2007, 2010, and 2013, as well as a coring expedition in 2010. Seismic analyses are needed to scan the sediment underneath the lake. The “air-gun” system, which produces seismic reflection data, is akin to ultrasounds often used with pregnant women. It allows us to detect the thickness of the mud that has accumulated on top of the lake’s bedrock for almost one million years. Using these data, we were able to identify potential sites for coring/drilling. The 2010 coring expedition gave us several piston cores from Lake Towuti. Each was about twelve meters long, and their base was dated to approximately 60,000 years ago. We have done various physical, biological, and chemical analyses on these cores, and our results suggest major environmental and climatic changes in the region during that period. These results confirmed our conviction that Lake Towuti is indeed a great site to study the climatic and environmental changes of Indonesia. Now, with our newly recovered cores from the TDP, we hope to capture and understand the major changes that might have occurred further back in time.

E T
In his day, all Alfred Russel Wallace needed was a letter from the Dutch in order to collect his specimens in Nusantara; over 150 years later, your legal requirements are much more substantial. Also, the setup for such a drilling project is extremely complicated. Can you give us a sense of the logistics involved in doing this kind of research?

S W
As the TDP is an international effort, up to twenty-five foreign researchers were involved in the actual drilling operation. We had to complete plenty of paperwork for various governmental institutions and at different levels of government. We also brought in multiple containers and major drilling and research equipment from the U.S. and Germany. These containers and the drilling rig are quite big, and there were only a few vehicles and cranes that could be used to transport them to our dock once they arrived in Sulawesi. We had to coordinate with various shipping and transportation contractors to help with the movement of our equipment. Due to technical difficulties in the middle of our project, the project had to be extended by about two weeks, and we shipped in spare parts for our hydraulic pumps from the U.S. They were not easy to locate and eventually arrived safely in Sulawesi, but the journey was not without hiccups. Finally, the delay in our departure from Sulawesi caused us some problems, as it forced us to reschedule the demobilization of our equipment and rebook our return tickets. Neither task was easy given that our final departure date was so close to Idul Fitri, a major holiday in Indonesia.

A S
You are not the only scientist presently working at the Towuti Drilling site. Can you tell us about your team? What scientific fields are your colleagues working in, and what they are looking for? More specifically, what it is that you are hoping most to find?

S W
There are about thirty scientists working either at the site or the on-shore laboratory during the drilling project. The scientists, led by Dr. Russell and Prof. Bijaksana, come from five countries and have different scientific backgrounds. Some of us, like myself, are interested in climate questions. How has rainfall history evolved since the lake was formed? What caused rainfall patterns to vary? What was the environmental response to changes in climate over time? We are using various proxy data extracted from the sediment cores to answer these questions. A few scientists involved in TDP are interested in the geological processes that formed the lake. They want to determine the age of the lake, as well as the evolution of its physical form over time. For instance, we saw a sequence of soil, river sediment, peat, and lake sediment in some of our drilling sites, therefore allowing us to deduce the geological evolution of the lake, as well as how the water level might have varied over time. We also found several thick volcanic ash layers, which might have originated from volcanic eruptions in northern Sulawesi. If these layers can be tied to previously identified volcanic ash layers from the region, they can be used as time-horizon markers. Lake Towuti also hosts many endemic species and is surrounded by one of the most diverse rainforests on Earth. The biologists among us are especially interested in finding microfossils from the sediment, which may help them understand the rates of biological evolution, as well as the sensitivity and resilience of Towuti’s aquatic and rainforest ecosystems to climatic and environmental changes. Finally, several biogeochemists are also involved in this project. Lake Towuti is among the world’s largest iron-rich lakes. The ophiolite bedrock surrounding the lake supplies metals that drive important biogeochemical processes. Drill cores allow us to expand our understanding of these processes and closely study the impact of climate changes on environmental chemistry. By doing so, we hope to improve our knowledge on how to maintain a sustainable Towuti ecosystem in light of recent changes in climate and human land-use.

Fig. 01. Core drilling rig in Lake Towuti. Photo courtesy of the TDP.

A S
Given the size of this project, and the remote area you are working in, it must attract a lot of attention. Are there any concerns by area residents about the project? What do the local communities think about your research?

S W
We have received a lot of attention not only from national and local media, but also from those living near the lake. The locals were particularly surprised to see huge cranes and other heavy equipment being transported through their villages to the project’s dock, where the gear was assembled into a working drilling barge. The confused locals thought that this equipment was owned by PT Vale Indonesia, a nickel-mining company operating nearby, which has often had a contentious relationship with the locals. Unfortunately for us, the words “drilling” and “project” also carry with them negative connotations. Within days after the project began, false rumors about our intentions started to spread widely. Some people that we met thought that we were going to do oil and gas exploration. Others were deeply concerned about the potential environmental impacts of the project; many asked us if a Lapindo-like mudflow disaster could happen because of our project.¹ The culmination was a demonstration staged by approximately 100 local residents and NGOs two weeks into the project. We listened to them, answered their questions, and worked hard to explain what the TDP really is. We were quite surprised with the moves against our scientific project, as Dr. Russell and Prof. Bijaksana had already held a meeting with the subdistrict head and local leaders two months prior to the start of the drilling. In addition, the district chief had known about this project since 2012, when it was still in the planning stage, but the information did not trickle down to area residents. The fact that 2015 was a busy year in politics, with elections of the district chief and village leaders looming, also added another complex dimension. We had already planned on doing outreach during our stay in Sulawesi, but the miscommunication with area residents forced us to go all out and work harder in conducting such activities. We held many meetings with various stakeholders: local leaders, government officials, local house of representative members, as well as representatives from NGOs. We also visited many elementary and high schools. In addition, we facilitated visits to our barge by government officials and locals alike, including school children, so that they may have a better understanding about our work. By the end, we were elated to learn that most locals had favorable views toward our project; many of them even expressed their open support.

E T
Regarding the drilling process itself, how long are the individual cores you are extracting from the lake bed? What condition do you extract them in (wet/dry, hard/soft, etc.)? And what will you do with this material next in order to study the samples collected?

S W
We hope to drill the sediment all the way down to the top of the bedrock. Lake Towuti’s water depth is approximately 200 meters, and the sediment underlying the water and overlying the bedrock has a depth of about 180 meters. To perform the drilling, we have to connect a bunch of five-meter-long drilling rods or pipes. When the actual drilling or coring equipment is lowered down through the inside of these rods, it brings along with it a three-meter section of plastic core liner. The drilled core sediment is captured in the core liner, and is prevented from falling back down by a core catcher when lifted up. Once the liner is out on the drilling barge, scientists cut it into smaller pieces (maximum length: 1.5 meters), put end-caps on both ends of the liners, and label them. Visual inspection suggests that the drilled wet sediments inside the core liners consist of different facies (for example bodies of rock with specified characteristics representing a certain  type of depositional environment), just as we had predicted before. We are able to see claydominated materials, sand-dominated materials, gravel-dominated layers, peat layers, as well as tephras (volcanic ash layers). The labeled sediment cores are then brought to the shore during our shift change, which happens twice a day. The cores are logged for magnetic susceptibility and other physical characteristics using a logger that we brought from Brown University and then temporarily stored in our field laboratory. We also sieve the sediment left in the core catcher and analyze these tiny samples under a microscope. The cores are currently on their way to the U.S. National Lacustrine Core Repository in Minnesota, where the scientists involved in the project will split them open into two halves during the core-splitting and sampling party later this year. The working halves will be photographed and logged for various physical characteristics using more sophisticated equipment and finally sub-sampled to be analyzed further by different scientists in their respective laboratories. Meanwhile, the archived cores will stay untouched in the repository in Minnesota.

E T
In our correspondence and previous meetings, you mentioned that the core samples you collect from Lake Towuti will provide up to 800,000 years of climate history. Can you explain more specifically how this climate knowledge is extracted from the sample? What sciences are involved in such processes of paleoclimatology?

S W
Paleoclimatology is the study of changes in climate taken on the scale of the entire Earth’s history. Humans started systematically collecting rainfall data from rain gauges and temperature data from thermometers only in the past few centuries. To obtain climate data from ancient times, we can use a variety of proxy methods borrowed from the Earth and Life Sciences that can tell us the amount of rainfall or temperature over a certain region during a certain period in the past. The physical, biological, and chemical characteristics of Lake Towuti’s sediment offer us clues of ancient climate history, which may help us understand the overall pattern of climate change and assist us in predicting how the climate might change in the future. For example, clay mineralogy may inform us of the types of materials that eroded easily and entered the lake, especially when the rainfall amount was high. During periods when the titanium level in Lake Towuti was higher than usual, for instance, we may deduce that those were relatively wet periods. Using pollens extracted from the sediment, then observed and counted under the microscope, we can learn about the types of plants that were living in the lake’s catchment area. Based on our pollen and leaf-wax carbon isotope analyses on the twelve-meter sediment cores from the 2010 expedition, we know that savannah was the dominant ecosystem in the area surrounding the lake between 33,000 and 16,000 years ago, during the last ice age. This indicates that the climate was much drier at that time. We hope to obtain environmental and climatic information further back in time using the longer drill cores from the TDP. One of our goals is to test whether the climate in the region was also dry during previous ice ages, and if so, why.

CORE SCANS OF LAKE TOWUTI
by James Russell

Lake Towuti began to form about one million years ago, when movements of Asia, Australia, and the Philippines reated fractures in the Earth’s crust. Movement long these fractures allowed the region of Towuti to sink, creating the 200-meter-deep depression the lake occupies today. During its initial stages of formation, lake Towuti was much smaller than today, and oftentimes existed not as a lake but a swamp, as depicted in the formation of peat, the sediment composed by remains of grasses, trees, and other plants. This contrasts dramatically with the green, banded clay that form in the deep lake today, where the sediment is composed of fine mineral material washed in from the soils and the remains of algae and other micro-organisms that live in the lake. Indonesia has been an extremely active region geologically throughout the past million years, with frequent volcanic eruptions that spew ash into the atmosphere. Much of this ash falls onto Lake Towuti and is buried in its sediments. Although these eruptions are often very destructive, they release nutrients into the water, stimulating biological productivity by diatoms and other algae. Reproduced here are four core scans we’ve produced at our labs since the expedition in 2015. In all, we drilled ten boreholes, which are anywhere from about 40 to 180 meters long, and cores from each hole are sectioned into 0.1-to-1.5-meter-long parts. This equates to over 1,200 core sections, each of which has a separate image, so this is just a very small subset of the images we have available. When viewed together, the core scans connote large changes over immense time-scales. Sediments vary from dark to light, red to green, and each of these colors is code for large environmental perturbations. In particular, large shifts between red sediments and green sediments record large changes in the lake level, driven by wet-dry cycles in regional climate that appear linked to the Earth’s ice-age cycles. These lake-level shifts cause large changes within the lake ecosystem and surrounding landscape. Even within each individual core section, there can also be visually interesting features, including aminations that record short-term rainfall and drought cycles. Finally, if we move beyond the macro-scale scanned images, we have microscopic images of fossil pollen, fossil diatoms, minerals, and ther materials from these cores, which tell us the composition of the sediments and can also be quite visually arresting.

Fig. 02. Peat shows deposits from the early stages of the formation of Lake Towuti, when it was not a lake but a swamp. All scans courtesy of James Russell.

A S
In their most cited paper on the Anthropocene, British scientists Simon Lewis and Mark Maslin emphasize the relationship between time and the Earth’s states.² How does it feel to think in such immense categories of time, nearly one million years into the past? Do you ever imagine the site of your research in that prehistoric state? What might it have looked like?

Fig. 03. Diatoms represent algal blooms in the lake that form
when nutrients from volcanic ash are released into the water.

S W
In the grand scheme of geological time scales, the past 800,000 years is actually rather short. Scientists believe that the Earth was formed approximately 4.6 billion years ago. Earth scientists like me are used to touching and observing rocks and sediment samples that are millions of years old. To us, they are not merely relics of  the past; they may also hold important scientific information about the present and the future, and as such are immensely valuable. For example, our study allows us to understand the natural variation of the environment and climate before humans began to populate the region.
    The present-day lake must have undergone significant changes in terms of size and geochemistry due to geological and climatic changes. Based on our preliminary data from the field we now know that lake levels have varied. Drilling at some of our sites yielded some non-lake sections, such as river and peat units. As mentioned, the landscapes around Towuti have not always been filled by lush rainforests. During dry and more seasonal periods, we might have seen savannah-type ecosystems instead. You may picture a vast expanse of grasslands dotted with some trees, similar to present- day Nusa Tenggara [Lesser Sunda Islands], during dry periods. At the same time, the lake level must have been lower, and the lake was smaller. It was a very different picture indeed.

E T
Do you expect that the team might find evidence in these samples for a contribution to the Anthropocene debate? According to the Geologic Time Scale, the Holocene Epoch begins in 11,650 BC, precisely the timescale you are investigating. Is there any possibility that you could find evidence of anthropogenic ecological change in your samples? Or, perhaps even a Global Stratotype Section & Point (GSSA) demarcation candidate for the Anthropocene?

S W
Given the slow natural sedimentation rate in the lake (averaging 0.02 cm/year), it is a little difficult to find evidence of anthropogenic ecological change in our samples. Archaeological and pollen-based evidence suggest that there was little disturbance by humans until the last century, when humans started to actively alter the landscape around the lake. I believe that the deforestation rate around Lake Towuti has increased significantly in the last decade. Not only has logging for timber become more intensive, but land clearance for pepper farming is also increasingly widespread. Maybe in a few more decades the evidence of these activities will become clear in the sediment.

Fig. 04. Green clay shows darker-layered sediment when the lake
level is high and the climate is wet.

A S
You also mentioned the relationship of Lake Towuti and the Wallace Line when you told us that you were researching some specific “climate refugia” in the area. Can you tell us about these refugia in relation to Wallace’s research on evolution, speciation, and biogeography? It would also be interesting to learn more about the term “climate refugium” itself.

S W
In a paper that I published recently,³ I concluded that elevation played an important role in determining the ecological manifestation of climate change in Indonesia. Based on two sets of climate and environmental proxy data from Lake Towuti and another lake located at a higher elevation (Lake Matano), there was a substantial difference in terms of rainfall and plant ecosystems surrounding the two lakes in the past. During the last ice age, when the climate in central Indonesia was generally drier and more seasonal, higher-elevation regions such as Lake Matano’s catchment were wetter and had a higher percentage of rainforest taxa compared to lower elevation regions.
    Since those high-elevation regions served as a home for rainforests amidst grassland expansion due to the drying climate, we may call them “climate refugia.” When the climate became more favorable (i.e. wetter), the rainforests might have expanded, encroaching on lower-elevation regions at the expense of the grasslands. Given that there are multiple highlands in Sulawesi, as well as other Indonesian islands, there might have been multiple “climate refugia” during dry periods.
    Our data point to several cycles of drying and wetting in the past, implying that a series of rainforest expansion and contraction cycles might have occurred around the refugia. The expansion of rainforests from nearby highlands could have played an important role in the evolution of plants. Different refugial peaks might have contained distinct plant species as they underwent different climatic and biological histories. Thus, when expansion occurred during wetter periods, and different yet still-related species intermingled, new plant species might have been produced. We thus hypothesize that the presence of climate refugia in Sulawesi played a crucial role in maintaining the high biodiversity in the region.

A S
We imagine that in 800,000 years there have been quite significant changes in Sulawesi and the ancient lake you are studying. Yet, we wonder if the more recent changes to Indonesia’s rainforests, biodiversity, and land-use are even more extreme, but on a much shorter timescale. To change the direction of our thought for a moment, what does the future look like given the current land-use practices in Indonesia? In another 100,000 years, what might one find if the Towuti Drilling Project occurred again at the same site?

S W
If you look at the landscape around Lake Towuti now, you will see patches of barren, cleared land. It isn’t difficult to pinpoint what the culprit was. Almost every day, and peaking on the weekends, you see smoke coming from the hills bordering the lake. Farmers want to clear the land to cultivate pepper, currently a very profitable crop, so they ignite the fires, essentially destroying all the trees in certain parcels of land. There are also illegal loggers who fell the trees around the lake. The loggers are slightly more environmentally friendly than the pepper farmers since they only selectively take down the big trees. These two activities are certainly worrying, given the potential environmental impacts on the lake ecosystem. It is also saddening given that Lake Towuti is actually a conservation area under the jurisdiction of the Ministry of Environment and Forestry.
    If one were to drill the lake sediment at the same site 100,000 years from now, one might encounter cores that have much higher sedimentation rates during the Anthropocene than the Holocene because of the increase in erosion. There is also a possibility that by then Lake Towuti will be much shallower than today. Chemical fertilizers that are used in pepper farming
may also increase the amount of nitrates and phosphates, thus increasing the biomass of phytoplankton and algae. As a result, sediment cores extracted in the future may have a substantially higher percentage of organic matter than what we found in 2015. Finally, we would see plenty of trash that humans threw in or near the lake. Plastic bags, bottles, and other relatively non-degradable materials could be easily seen in future sediment cores.

1 The Lapindo mud-flow disaster (also known as Lusi) refers to the continuous eruption of a mud volcano in Porong, Sidoarjo in East Java. This, the biggest mud volcano in the world has submerged a dozen villages, displaced some 30,000 residents, and prompted the  closure and route diversion of major highways since it started erupting in May 2006. There is an ongoing controversy surrounding
its trigger. Several quarters contend that the disaster is human-made, started by the blowout of a natural gas well drilled by PT Lapindo Brantas in the area.

2 Satrio Wicaksono, James Russell, and Satria Bijaksana, “Compound-Specific Stable Isotope Records of Vegetation and Hydrologic Change in Central Sulawesi Since 53,000 yr BP,” Palaeogeography, Palaeoclimatology, Palaeoecology 430 (2015): 47–56.

3 Simon Lewis and Mark Maslin, “Defining the Anthropocene,” Nature 519, no. 12 (March 2015): 171–80.