- <div style="background-image:url(/live/image/gid/32/width/1600/height/300/crop/1/41839_V14Cover_Lynch_Artwork.2.rev.1520229233.png)"/>
Breathtaking structures: Macrotermes michaelseni builds mounds featuring passive thermal regulation and air ventilation
Lake Forest College
Lake Forest, Illinois 60045
Despite having small bodies (only a few millimeters long) and featuring a fragile exoskeleton, termites are excellent builders of elaborate structures. These constructions – some of them reaching heights of 5 meters or more and entirely made of soil and termite saliva – are the result of the collective work of thousands of termites.  Several researchers are studying termite colonies and the mounds they build. Hunter King, a postdoctoral fellow in Applied Mathematics at the Harvard Paulson School of Applied Science and Engineering (SEAS) and his team were trying to find an answer to a biological mystery that has puzzled scientists for years. King explained that the termite mound represents a perfect example of construction built by decentralized swarm. Furthermore, he stated that this construction has a surprisingly clever way ofregulating thermal and respiratory conditions. Studying two termite species, the group of scientists uncovered the function of the termite mound system. 
Similar to other animals such as ants and bees, termites live in colonies and as such, their combined strength exceeds by far the power of an individual animal. The group emerges in a complex organization that displays properties not found in the individual termite. Together, termites can move hundreds of pounds of soil and tons of water. As Nikita Zachariah, a doctoral student at Indian Institute of Sciences explained, the termites build the mound by joining tiny rolled-up balls of soil. The insects fill their mouth with moist soil, mix it with their own saliva, which acts as a sort of cement, and transport the material to the construction site.  When referring to such a mound, scientists often speak of a so-called superorganism that demonstrates a life by its own. Interestingly, when looking at such a conglomeration of individuals, one does not find a single member who is in charge of the entire group. Not even the individual, which we as humans refer as the queen, represents the decision-making authority for the mound. The termites are blind and only react to local chemical traces placed by other termites and to differences in temperature, humidity, and airflow caused by environmental circumstances and conditions inside the mound. The mound – termite relationship represents a cause-reaction system: while the behavior of the termites shapes the structure of the mound, the mound itself also effects the behavior of the termites.  The colony takes notice and reacts to changes in food, weather conditions, and attacks by other animals. Although the individual termite has a rather short lifespan (when compared to most other animals), the colony can survive for several years, constantly evolving itself.
Inhabiting such a structure not only confronts the termites with problems of organizational nature (i.e. space constraints, accessibility and nutrition supply) but also of environmental nature (i.e.regulation of temperature, respiratory gas and moisture) to adequate levels. Until recently, scientists did not understand how exactly the geometry of the termite mound presents a foundation for the precise control of the internal climate.  King and his team studied the colony structures of the termite species Macrotermes michaelseni and Odontotermes obesus. The team was interested in understanding how the termites manage to control the climate of these microhabitats to optimize conditions for the termite society and for cultivating fungi.
Figure 1: Photo of a termite (M. michaelseni) outside the colony`s mound. 
While M. michaelseni is found in the sub-Saharan region of Africa, O. obsesus inhabits regions of India and East Asia. M. michaelseni builds mounds that have a distinctive architecture, divided into three main components: a mound that is cone-shaped, a tall and thin spire tilted at an angle that is comparable to the sun`s average zenith angle, and a broad pediment. The mound features a complex system of tunnels that are grouped in three categories: a central tunnel acting as a chimney from the ground to the top of the mound, vertical-based conduits that lay a few centimeters below surface level, and a network of lateral tunnels that act as connections between the chimney and the surface conduits. 
Although O. obsesus`s architectural structure also ends in a peak, it forms an overall structure that differs from that built by M. michaelseni. O. obsesus builds mounds that feature flute-like structures on the outside, making the entire structure resemble a cathedral. Figure 1 illustrates the colony structures of the two termite species.
Figure 2: Mound built by the termite species O. obesus (upper A) and mound built by M. michaelseni (lower A). Filling the mound with gypsum and removing the material reveals the interior network of conduits in both structures (O. obesus, D, M. michaelseni, B). ,
Like other mound-building termites, both species cultivate a fungus garden. They have a symbiotic relationship with the basidiomycete fungi of the genus Termitomyces;while the termite colony creates ideal growing conditions in terms of temperature, humidity, and protection, the fungi provides food (in the form of sugars) for the colony. Studies have determined that the cultivation of the fungi garden represents monocultures that are evolutionary stable and are used to decomposed cellulose material. 
Figure 3: M. michaelseni cultivating a fungus garden. 
The fungus garden and the living structures (the nest) are not situated in the mound itself, but underneath it. Aside from being an entry for construction and repair, the mound has a relatively inhabited existence. This raises the question of why termites would build such an immense structure without living in it? Scientists hypothesized that the structure`s main purpose is to control the environmental factors within the colony itself by regulating temperature, humidity, and gas concentrations. The mounds built by M. michaelseni, feature a network of pathways organized to capture the energy dissipated as a by-product of the termites’ metabolism: heat. Around the 1960s it was hypothesized that the internal heat within mounds of Macrotermes natalensis arising from the termites’ metabolic processes would drive a convective air flow with the purpose of regulating temperature and ensure gas exchange within the structure.  Later studies conducted on Macrotermes bellicosus led to the assumption that thermoregulation is not only guided by the mound’s internal conditions but also by external solar heating. 
The team around Hunter King worked with Scott Turner, an American physiologist at SUNY College of Environmental Science and Forestry, who specializes in animal-built structures. Together, they wanted to understand how the termite mound functions and what consequences its shape has.  Hunter et al. points out that it is important to note that past studies examined airflow velocities based on measurements of indirect quantities such as temperature and gas concentrations. In contrast, Hunter et al. built a sensor system to directly measure the airflow within a mound.  The process of acquiring data was a challenge. First, since the air in the mound flows very slowly, commercially available measurement devices were not well-suited for the study. Second, King explained that not only the complex chamber geometries presented a challenge, but also the fact that any intrusion into the mound would alarm the worker termites who would rush to the affected region and either try to attack the invading object or spread a gluey and corrosive saliva on the sensors. Therefore, the team had to custom-build the sensors, keeping in mind that they not only had to be calibrated to the special geometries of the mound but also had to be cheap to compensate for eventual loses. 
Inhabiting the southern African savanna, the termite M. michaelseni experiences direct sun exposure with temperature differences of ΔT≈15–20°C and winds of up to 5m/s. King et al. considered two possible causes: fluctuations of external wind and thermal gradients within the termite mound. In addition, the team measured the daily CO2 concentration. The scientists performed their study in the Cheetah Conservation Fund in Otjiwarongo, Namibia, and its duration was about 1 month during two consecutive years. To measure the internal air flow King et al. inserted a sensor system featuring a central heating component into the vertical surface conduits located about 1m above the ground. The relative magnitudes in temperature peaks were recorded by the neighboring components located next to the heating component. In conjunction with measuring each internal flow, the scientists recorded the wind speed and direction with a cup-and-vane anemometer at the height of 2 m above ground level and approximately 3 m from the mound. To measure temperatures inside the mound King et al. placed a total of 8 temperature and humidity data loggers at a height of approximately 1 m in the nest (approx. 0.5 m below the surface, at the center and at the top within the termite mound. The scientists compared their data with the measurements obtained during a 1-year cycle. They quantified CO2 concentration by inserting tubes with an inner diameter of 6 mm at various places into the mound reaching the nest and the center of the chimney. The tubes were left overnight for the termites to seal the gaps between the tubes and the drilled holes. To the surprise of the researchers, the tube endings inside were not sealed by the termites. The team withdrew about 500ml of air every 15min for a period of 24h using a syringe for each tube, thus channeling air through a chamber featuring a CO2 sensor.
King et al. noted that the thermal gradient within the termite mound is maintained even though CO2 is well mixed through it. Furthermore, their findings show that significant fluctuating flows are present in mounds of M. michaelseni. Despite the variations in flow speed, the scientists were able to draw a robust trend: they concluded that at night, there is a strong air flow down the conduits, while during the day the air flow changes its direction. Furthermore, the pattern of internal temperatures allowed the scientists to construct azimuthal dependences following the path of the sun: eastern conduits warm first in the morning, followed by northern and western conduits, while the southern conduits remain the coolest region. It is possible that conduits located in shaded regions resulting in a temperature gradient within the mound are involved in the creation of a downward flow.
Figure 4: Change in internal temperature gradients in a cycle of 24 hours. The east side of the mound reaches its warmest point during the morning when the sun rises. The west side reaches its warmest point during the time when the sun sets. Since the termite habitats are located in the southern hemisphere, the north side is more exposed to direct sunlight than the south side resulting in a higher surface temperature. 
Looking at CO2 levels within the mound, King et al. found no large gradients of CO2. They state that the CO2 concentration was around 5% throughout the measurement period. Other mounds in the region featured similar CO2 levels ranging from 4 to 6%.
Figure 3: Change in CO2 concentration in the nest and in the chimney during a 24h cycle. The team of scientist measured CO2 levels both in the belowground nest and directly above the nest, approx. 1 and 2 m from the ground. The relatively steady and uniform concentrations indicate than there is almost none internal mixing. The inset in the right bottom corner represents corresponding data from the nest and mid-chimney of a mound by O. obesus in India. In both measurement procedure, the sensor error was at most ±0.18%. 
King et al.`s study revealed that neither wind nor the metabolism plays a sole role for internal transient flows of air. The scientists stated that there is a poor correlation between wind and the internal thermal difference and that daily occurring temperature changes are the main reason for convection flows ventilating the structure. Since temperature oscillations are occurring across our planet, the exchange mechanism is only dependent on the characteristics of the mound structure. King et al.’s study showed that air within the mounds of M. michaelseni flows as a convective cell guided by a diurnally oscillating thermal schedule. Figure 4 shows a schematic of the airflow through a night-day cycle inside a termite mound.
Figure 4: Schematic (top and side) of the thermal gradient (indicated with colors of purple and red) and airflow directions (indicated by arrows) inside mound during a night-day cycle. During the daytime (nighttime), the periphery is hotter (colder), while the center is cool (warm). This condition creates a convection cell with upwards (downwards) flow in the periphery and downwards (upwards) flow in the center. During mornings, the sun directly heats the east side of the mound, so that it is hotter than the west side; during evenings, the west is hotter than the east side. Furthermore, since mounds are located in the southern hemisphere, the north side of the mound receives more sunlight than the south side. It is important to note that due to measuring constrains, measurements of air velocity deep inside the mound were not possible to obtain. Therefore, the exact boundary between upwards and downwards flow remains known. 
While comparing their results with the findings from studies done on another termite species, O. obesus, the researchers saw similarities in the reparative mechanism of the mounds. However, they also found a few exceptions: in addition to the radial thermal gradient caused by daily temperature fluctuations, direct sunlight exposure causes angular thermal gradients. Furthermore, CO2 levels are around 5% inside the mound throughout the day. At night, the occurring airflow pulls the gas to the exterior wall where it diffuses to the external environment.  Comparing the mound architecture of the two termite species, King et al. noted that while O. obesus builds structures that have a greater radiator-like shape in regions that are less prominent to thermal oscillations, suggesting that the two conditions result in enhanced thermal gradients. The team of scientists also observed that the mounds of O. obesus featured thin channels located on the outer surface of the mound. These channels heat up rapidly compared to the tunnels inside the mound. And cause air to circulate: during the day air moves up the outer channels and down in the inside. During the night, the airflow changes its direction. Figure 5 shows pictures of the mound of O. obesus acquired using thermal cameras. The left half of the mound shows temperature distribution during nighttime while the other half shows the thermal gradients during the day. Grasping this process, the termite mound, built by swarm intelligence and collective behavior, represents a vital “organ” of the colony. Lakshminarayanan Mahadevan, Professor of Physics at Harvard University, states that “… the termite mound works like a slowly breathing lung, flushing CO2 out once a day and aerating the mound…” 
Figure 5: Picture of a mound built by O. obesus acquired using thermal cameras. The left half of the mound shows temperature distribution during nighttime, revealing the cool air within the outer channel structures as well as a hot mound centerpiece. The other half shows the thermal gradients during the day where the center of the mound is cooler than the outer vertical channels. 
The study by King et al. revealed that, in addition to their immense size and complexity (keeping in mind that they are constructed by such tiny creatures) they represent inspiring, passively ventilated structures built by animals. As such, the mounds can be seen as a functioning external lung of the colony. As King et al. stated, the ventilation of the mounds of the two studied termite species is a result of solar heating (direct and indirect sunlight exposure) as well as of thermal lag. The combination of the two mentioned factors lead to the creation of convective airflows between the interior of the mound and its periphery. The findings of the researchers not only give valuable insights and further understanding of the lives of termite colonies but may also have implications for engineering thermodynamics and for the design of thermal oscillation ventilation systems for buildings.  Together with Mahadevan and other researchers, King plans to study other termite species to find out if the thermal regulation and gas exchange mechanism is generalizable. In addition, they hope that their discoveries will inspire architects and civil engineers to apply these concepts of passive cooling and ventilation to infrastructure built for humans. 
 King et al. “Solar-powered ventilation of African termite mounds.” Journal of Experimental Biology, 2017 http://jeb.biologists.org/content/220/18/3260.long
 King, H., Ocko, S., Mahadevan, L. “Termite mounds harness diurnal temperature oscillations for ventilation.» Proceedings oft he National Academy of Sciences oft he United States of America (PNAS), 15 September 2015. https://www.pnas.org/content/pnas/112/37/11589.full.pdf
 Termite mounds could inspire new ideas for sustainable building ventilation. The Wyss Institute for Biologically Inspired Engineering at Harvard University, 31 August 2015 https://wyss.harvard.edu/termite-mounds-could-inspire-new-ideas-for-sustainable-building-ventilation/
 Turner, J. S. “Architecture and morphogenesis in the mound of Macrotermes michaelseni (Sjöstedt) (Isoptera: Termitidae, Macrotermitinae) in northern Namibia.” Cimbebasia, 2000. https://www.esf.edu/efb/turner/publication%20pdfs/Cimbebasia%2016_00.pdf
 Wageningen University and Research Centre. “Termites create sustainable monoculture fungus farming.” ScienceDaily, 22 November 2009. https://www.sciencedaily.com/releases/2009/11/091120000437.htm
 Wilkinson, R. “Namibia`s Castles of Sand.” Web blog, 30 June 2010. http://www.wilkinsonsworld.com/tag/macrotermes-michaelseni/
 Harvard University. “Taking apart termite mounds: Study reveals how distinctive mounds are ventilated, could offer lessons to human architects.” ScienceDaily, 2 September 2015. www.sciencedaily.com/releases/2015/09/150902155635.htm
 Burbeck, S. “Example mergent Phenomena” Web blog, Evolution of Computing, 22 June 2018. http://evolutionofcomputing.org/Multicellular/Emergence.html
 American Institute of Physics (AIP). “Climate control in termite mounds.” ScienceDaily, 25 November 2014. www.sciencedaily.com/releases/2014/11/141125074646.htm
 Nguyen, S. “Termites` Cathedrl Mounds” Harvard Magazine, November-December 2015. https://www.harvardmagazine.com/2015/11/termites-cathedral-mounds
 Pendharkar, V.. “A Method in the Madness: How Termites Build and Repair Their Mounds” The Wire, 28 July 2017 https://thewire.in/science/termites-mound-bolus-granular-hydrogel
Eukaryon is published by students at Lake Forest College, who are solely responsible for its content. The views expressed in Eukaryon do not necessarily reflect those of the College.
Articles published within Eukaryon should not be cited in bibliographies. Material contained herein should be treated as personal communication and should be cited as such only with the consent of the author.