A microscopic plankton shell can be weighed without touching it—and potentially with better repeatability than an ultra-microbalance. This study uses X-ray microcomputed tomography to turn imaging into precise bulk-density measurement. X-ray microcomputed tomography is becoming an important technique for examining the precise morphometry of small objects, because it enables nondestructive, highly accurate measurements at micrometer-order resolution.

That makes microcomputed tomography useful in Earth sciences for clarifying how genetic associations and the surrounding environment affect the morphology of micro-sized organisms. The gap is that the actual analytical methods and the points needed to produce reliable data have rarely been discussed in detail.

Using Globorotalia inflata, the study describes precise microcomputed-tomography morphometry, demonstrates long-term stability, and proposes test bulk-density calibration with artificial carbonate phantoms. Three-dimensional morphometry with microcomputed tomography has the potential to solve these problems.

Microcomputed tomography has been widely discussed for estimating carbonate solubility and the calcification capacity of marine calcareous materials. It has also been used in attempts to estimate carbonate-ion concentration in past seawater, habitat depth, and biological effects of anthropogenic environmental changes since the Industrial Revolution.

The methodology, accuracy, and reproducibility of microcomputed-tomography densitometry for microfossils have not yet been discussed in detail. The study therefore describes microcomputed-tomography measurements and the precautions that must be considered.

It also examines measurement stability and develops and applies new reference materials for calcium-carbonate density measurement. The target was a single test of the planktonic foraminifer Globorotalia inflata, recovered from water pumped from the surface of the Kuroshio Current in the western North Pacific.

The reference material was a single grain of limestone crystal formed from stable oxygen and carbon isotopes, identified as NBS19, or NIST RM8544. The reference material was analyzed at the same time as the target specimen, allowing later adjustment for energy fluctuations in the X-ray radiation from the X-ray tube.

To prepare the standard, the limestone grain was gently dissolved in zero point zero one molar hydrochloric acid, molded into a spherical-to-subspherical shape, and used to average the X-ray transmission distance. With white X-rays, high-energy X-rays preferentially penetrate the object, while low-energy X-rays are absorbed at outer or internal surfaces and reach the detector only after attenuation.

This selective absorption is called beam hardening, and it can adversely affect density measurements of microfossils made with microcomputed tomography. A metal filter can reduce beam-hardening artifacts, and a simulation found the greatest suppression below twenty kilovolts with two-hundred-micrometer-thick, high-grade aluminum.

The study used that two-hundred-micrometer aluminum filter in front of the X-ray detector for all test bulk-density measurements. The same concept applies to calculating the test bulk density of a foraminiferal sample. Attenuation coefficient and gray value are not exactly the same physical quantity, but both are related to X-ray attenuation, so their relative relationship holds.

They can therefore be made dimensionless and expressed as relative density by normalizing with the gray value of the standard sample. The preparation method produced carbonate phantoms with various porosities, or bulk densities, because different pressures created different states of crimping between calcium-carbonate particles.

The phantoms and a single NBS19 grain were analyzed by microcomputed tomography under the same conditions used for the later foraminiferal-test analysis, to develop a calibration equation between test bulk density and CT number. Figure one illustrates beam hardening in a limestone reference sample.

Without a metal filter, panel A shows higher gray values around the edge and lower values toward the center, while panel B, acquired with a zero point two millimeter aluminum filter, appears more homogeneous. The corresponding gray-value distributions in panels C and D report skewness values of minus three point two and two point five, respectively.

This matters because the filter-based correction supports density analysis of foraminiferal tests with calcium carbonate transmission paths up to one hundred to three hundred micrometers. Figure four shows cross-sectional μCT images of three artificial calcite phantoms: P2, P4, and P8, pressed for ten minutes with weights of two, four, and eight tons.

Their displayed bulk densities are d equals one point eight two nine, one point nine nine, and two point zero nine seven, while the color scale maps lower to higher density. The visibly fairly uniform textures support the authors’ assessment that these phantoms were sufficiently homogeneous for calibration and for relating CT numbers to bulk density.

The cross-sectional images indicate that the artificial carbonate phantoms were sufficiently homogeneous for use as calibration phantoms. The relationship between bulk density and CT number is shown in Figure 5, with CT number measured ten times for each calibration phantom and average value and relative standard deviation calculated.

Figure five plots CT number against test bulk density for the carbonate phantoms P2, P4, and P8, alongside reference material NBS19. Each phantom was measured ten times, and the plotted points show a clear linear relationship, with the dotted regression line reporting an R-squared value of zero point nine nine nine four.

The gray band marks the ninety-five percent confidence interval, supporting the authors’ use of bulk density as a calibration parameter for CT measurements. Across the investigation, the average CT number was eight hundred fifty-four point nine, with a range from eight hundred thirty-nine point six to eight hundred seventy-one point zero.

The standard deviation was seven point seven, the variability was zero point ninety percent, and variability showed no clear trend over the two-month investigation period. Figure six plots CT number across thirty-one analyses conducted from the first of October to the thirtieth of November, twenty twenty-two, with test bulk density indicated on the right-hand scale.

The points cluster tightly around the reported average CT number of eight hundred fifty-four point nine, spanning eight hundred thirty-nine point six to eight hundred seventy-one point zero. This compact distribution, with no clear time-related trend, supports the authors’ conclusion that the CT measurement was stable over the two-month investigation.

Using Equation four, the mean test bulk density of Globorotalia inflata was two point three plus or minus zero point zero two grams per cubic centimeter. The weight of a single test calculated from CT number was twenty-one point eight plus or minus zero point two eight micrograms.

The bulk-density value was close to the average reported for several planktonic-foraminifera species in the western North Pacific, and the weight was within the reported range for individual Globorotalia inflata in modern water. Table three compiles thirty-one micro-CT examinations of Globorotalia inflata shells across the two-month investigation, linking scan settings with density, volume, surface area, thickness, and calculated weight.

The mean CT number is 854.9, with values from 839.6 to 871.0 and variability of 0.90 percent, while the mean detector-responsiveness value, k, is 3.4. This matters because the table shows both the measured shell properties and the scan-to-scan variation used to assess their stability.

The CT number for the foraminiferal test was very stable over two months: variability remained within zero point nine percent, with no systematic fluctuations. Volume and surface area each had reproducibility of one point five percent, while calculated weight averaged twenty-one point eight micrograms with a standard deviation of zero point two eight micrograms.

The weight variability was comparable to, or slightly better than, values reported by others for planktonic foraminifera, so weights can be obtained with comparable accuracy without an electronic microbalance. The electronic ultra-microbalance had variability of eight point two percent, much greater than the one point three percent obtained with microcomputed tomography, supporting the reliability and objectivity of the microcomputed-tomography approach.

There are important points to note when test bulk density is used as an index of carbonate dissolution. In strongly dissolved, two-layered foraminiferal tests, the layer most vulnerable to dissolution can be removed first, leaving the less-vulnerable layer behind.

As a result, a strongly dissolved test could still have a high CT number and lead to the erroneous conclusion that it has not undergone dissolution. CT number should therefore be applied to species with a known dissolution process and considered together with microstructure revealed by SEM, CT images, or both.

The CT numbers of the novel artificial carbonate phantoms showed excellent linearity with independently measured bulk density, indicating their potential as reference materials for measuring marine-calcifier test bulk density. Calibrating CT numbers with these phantoms is expected to help standardize measurement results between laboratories.

The study shows that calibrated microcomputed tomography can measure foraminiferal test bulk density and weight with strong stability, while also warning that density alone cannot diagnose dissolution without seeing the shell’s microstructure.