Closure to “Axisymmetric Simulations of Cone Penetration in Biocemented Sands”

Abstract

The original paper presented a numerical study of cone penetration in cemented sands. Although microbially induced calcite precipitation (MICP) shows promise as a ground improvement technique through extensive laboratory studies (e.g., Montoya and DeJong 2015; Lee et al. 2022), there is no cone penetration test (CPT)-based method for evaluating the increase of cementa-tion due to MICP. The objective of the study was to develop a relationship between change in cone tip resistance (Δq c) and the "apparent" cohesion (c) attributed to MICP treatment. The primary points raised by the discusser were the actual difficulty in sampling cemented sands, terminology of characterizing cohesion from MICP as "apparent" cohesion, interpretation of some of the data from published sources for use in the study, and application of the developed relationships between "apparent" cohesion and normalized shear wave velocity to example data, and how cementation levels are classified. Additionally, the discusser raised several practical considerations for MICP treatment application for liquefaction mitigation. The writers of the original paper provide responses to the discussion in the following. The discusser provided several examples of projects where intact samples of cemented sands were successfully obtained for laboratory testing. These examples illustrate that intact sampling of cemented sands is feasible. The writers agree with the discusser on the overall description of the mechanical behavior of cemented sand and the importance of obtaining high-quality samples. Several of the writers have compiled a comprehensive database on the mechanical properties of naturally cemented and biocemented sands (the focus of this research), which is currently under review. Based on the extensive review of available test data on biocemented sands, the writers deduced that the shape of the stress-strain response is not an indicator of the quality of the sample. An indication of good sample quality might be, as the discusser mentions, if the retrieved naturally cemented sample is reproduced artificially in the lab, tested under similar conditions, and yields similar results. Another possible indicator of good sample quality might be measuring a shear wave velocity in the laboratory sample that is comparable to that measured in situ. Otherwise, the shape of the stress-strain response can be representative of some cementation level but not necessarily the intact cementation level because ce-mentation bonds tend to be brittle. The ability to obtain quality samples decreases as the cementation level becomes lighter, and sampling of soils with only a few percent cementation (by mass) is very difficult from the writers' experiences. Cone penetration testing, informed by the paper under discussion, with geo-physical testing and soil sampling for MICP-treated sites can be a basis for evaluating the changes in mechanical soil properties and the variability of these changes across a treatment zone. Cohesion for MICP-Cemented Sands Many of the discusser's comments were related to the cohesion in cemented sands and the development of Fig. 4 of the original paper. The discusser was concerned with the use of the term "apparent" for cohesion in cemented sands. The term was used in this paper in the context of cementation to indicate the probable nature of the additional cohesion coming from the cementation. Research on cemented sands continues to mature, and the complexities introduced require more comprehensive work before the depth of understanding is on par with clean sands and clays. Recognizing this, the writers use the term "apparent cohesion" to indicate that the bonding capacity introduced by cementation can be reasonably represented using the cohesion term in the Mohr-Coulomb constitutive model. The writers thank the discusser for pointing out errors in Fig. 4 of the original paper. The writers confirm that the data from O'Donnell et al. (2017) were not included in the development of this relationship and the data points were mislabeled. The legend should only reference the triaxial tests from Nafisi et al. (2020) and Gomez et al. (2018). Corrections to the original paper can be found in the "Erratum" section of this closure. Both references reported shear wave velocity measurements during their experiments. The strength parameters (cohesion and friction angle) from Nafisi et al. (2020) were interpreted using three envelopes: linear, nonlinear, and bilinear. Therefore, three sets of strength parameters were reported by Nafisi et al. (2020). The data in Fig. 4 of the paper under discussion from Nafisi et al. (2020) corresponded to the cohesion from the linear failure envelope fitted to different soil types treated at light and moderate cementation levels. These values are © ASCE 07023015-1