Research ArticleAPPLIED SCIENCES AND ENGINEERING

The architectural design of smart ventilation and drainage systems in termite nests

See allHide authors and affiliations

Science Advances  22 Mar 2019:
Vol. 5, no. 3, eaat8520
DOI: 10.1126/sciadv.aat8520
  • Fig. 1 Termite nest locations, excavation, and x-ray tomographic imaging.

    (A) Two sampling locations, Nguekokh (Senegal) and Kankan (Guinea)—marked by yellow circles, were selected for this study. The Senegal nest was taken from a region that is sandy on the surface (brownish in the satellite map), whereas the Guinea sample was taken from a more vegetated region. The map in this image is taken from Google Maps, with the data provider listed at the base of the image (Imagery 2017 Landsat/Copernicus, Data Scripps Institution of Oceanography, National Oceanic and Atmospheric Administration, U.S. Navy, National Geospatial-Intelligence Agency, General Bathymetric Chart of the Oceans, Map data 2017 Google). (B and C) Nests in the field in Senegal (B) and Guinea (C). (D and E) These nests were excavated from the field and brought to the laboratory for x-ray imaging. The yellow dashed lines in (D) and (E) show the boundary between the upperground and the underground portions of the nests. [Photo credit for parts (B) to (E): Christian Jost]. (F and G) The excavated nests were imaged in three dimensions nondestructively with a medical x-ray tomography scanner at a pixel resolution of 0.3 to 0.6 mm. The imaging plane was clipped vertically to show the interiors of both nests. Red and yellow represent solid material and the inner channels of the nest, respectively. (H) Histograms of the thickness of the outer walls of the nests, which was measured from the tomographic images (F) and (G) in the upper parts of the nests that were exposed to the atmosphere.

  • Fig. 2 Thickness profiles of the inner solid walls and channels of the nest.

    (A and B) Thickness maps of the Senegal nest, showing the width of inner channels (A) and inner solid walls (B). The regions close to the outer wall were not considered in this analysis. (C and D) Thickness maps of the Guinea nest, showing the width of inner channels (C) and inner solid walls (D). The semitransparent box indicates the parts of the nest that were underground. (E and F) Histogram of thickness maps of the inner channels (E) and the inner walls (F).

  • Fig. 3 X-ray microtomography analysis of termite nests.

    (A and B) Two-dimensional grayscale cross sections of the x-ray microtomographic images of the outer wall of the Senegal (sample 1) (A) and Guinea (sample 1) (B) nests at a voxel size of 5 μm. Gray and white spots represent solid matrix and metallic elements [consistent with the XRD analysis], respectively, while black represents empty (void) space filled with air. In the Senegal nest (A), the smaller pores (intrapellet pore space) and the larger pores (interpellet pore space) are distinguishable, whereas the Guinea nest (B) shows only interpellet pore space, because of a larger fraction of clay indicated by shrinkage cracks in the solid matrix. (C) The solid part of the Senegal nest was dissolved in water to form a slurry in a glass vial, which was then air-dried without compaction, hereafter called Senegal random pack, and imaged with x-ray microtomography. (D to F) Three-dimensional images of the Senegal, Guinea, and Senegal random pack, with their color coding for pores, solids, and metallic elements. (G) Probability density function (PDF) and cumulative frequencies of the pore size of the samples shown in (D) to (F). We also show data (red) for a subset (SSc) taken from the Senegal nest sample in which the smaller pores were isolated for pore size comparison (refer to text and fig. S2 for further details). The vertical dashed gray line indicates the upper bound of the pore radii of smaller pores, obtained from subsets SSa-SSd. (H) Porosity-permeability relationship for the Senegal, Guinea, Senegal random pack, and four different subsets of isolated smaller pores of the Senegal nest. The error bars of the porosity are extracted from the porosity values of each slice along the z axis of the image.

  • Fig. 4 Flow fields and CO2 diffusive fluxes.

    (A and B) Flow field simulation using Navier-Stokes equation on the Senegal nest (sample 1), showing normalized pressure (A) and velocity (B) fields. The black arrow in (B) shows the inlet face and the direction used in all simulations (A to K). The boundary condition used in the flow simulations corresponds to a pressure gradient of 1 Pa/mm. A higher pressure is imposed on the inlet face of the sample. (C) CO2 flux visualization in the porous matrix of the Senegal sample, which was estimated by solving Fick’s second law. The boundary condition used in this simulation corresponds to a 5% change in CO2 concentration (relative to an atmospheric CO2 concentration of ~0.0164 mol/m3) across each 1 cm thickness of the nest wall. (D to I) Similarly, these simulations were conducted on the subsets of the Senegal nest containing smaller pores and the Guinea nest. Pressure (D), velocity (E), and CO2 flux (F) in the subset (SSc) containing only smaller pores in the Senegal nest. Pressure (G), velocity (H), and CO2 flux (I) in the Guinea nest (sample 1). Pressure (J) and velocity (K) fields in the Senegal random pack. (L) PDF of the logarithm of the pore-scale velocities in different samples.

  • Fig. 5 Heat flux simulations.

    Heat flux streamlines colored by the magnitude of the heat flux obtained for an applied temperature gradient of 1 K/cm in the Senegal nest (A), smaller pores in the Senegal nest (SSc) (B), and the Guinea nest (C). The black arrow in (A) shows the direction of temperature boundary condition used in all simulations (A to C). The light and dark gray in the solid matrix show sand grains and clay, respectively, whereas white represents metallic elements.

  • Fig. 6 Drainage of the Senegal nest.

    (A) A two-dimensional grayscale cross section of the dry Senegal sample (sample 2). Air is represented by black, and the solid phase is represented by gray. (B and C) The sample was saturated with 0.16 ml of potassium iodide (KI)–doped water (for improving x-ray absorption contrast, see Materials and Methods) from the top, which was then allowed to drain. The x-ray tomographic image represents the sample at approximately 2.5 hours (B). An axis connectivity analysis was performed on the air phase (C). For the connectivity analysis, the connected voxels belonging to the pore space that spanned across the full length of the sample (along an axis) were considered. Here, light blue represents the connected percolating air cluster spanning across the length of the sample. Various other colors show disconnected air clusters. Light green at the base could be connected; however, because of its presence at the border of the image, it is assigned as a disconnected phase during image processing. (D) Three-dimensional visualization of the percolating air cluster in the drained sample. (E and F) Flow field simulations on the air phase of the drained sample showing pressure (E) and velocity (F) fields. (G) Pore occupancy analysis of the drained sample. The plot shows the fraction of the pore space occupied by water [white in (B)], connected air cluster [light blue in (C) and (D)], and disconnected air [various colors in (C)]. (H) PDFs of the velocity fields for the dry and the drained Senegal sample. The drained sample shows a considerable amount of stagnant regions with flow focused through fewer open paths.

Supplementary Materials

  • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/5/3/eaat8520/DC1

    Air Percolation Analysis

    Fig. S1. XRD analysis of the nest material.

    Fig. S2. Subset selection for the smaller pores in the Senegal nest.

    Fig. S3. High-resolution x-ray microtomographic images.

    Fig. S4. REV analysis.

    Fig. S5. Four-phase separation of the Guinea nest sample.

    Fig. S6. Computation of percolation threshold.

    Fig. S7. Air percolation analysis in the outer wall of the termite nests.

    Table S1. Mineral compositions of the Senegal and Guinea nest material obtained from the XRD analysis.

    Reference (46)

  • Supplementary Materials

    This PDF file includes:

    • Air Percolation Analysis
    • Fig. S1. XRD analysis of the nest material.
    • Fig. S2. Subset selection for the smaller pores in the Senegal nest.
    • Fig. S3. High-resolution x-ray microtomographic images.
    • Fig. S4. REV analysis.
    • Fig. S5. Four-phase separation of the Guinea nest sample.
    • Fig. S6. Computation of percolation threshold.
    • Fig. S7. Air percolation analysis in the outer wall of the termite nests.
    • Table S1. Mineral compositions of the Senegal and Guinea nest material obtained from the XRD analysis.
    • Reference (46)

    Download PDF

    Files in this Data Supplement:

Stay Connected to Science Advances

Navigate This Article